Influence of polymer modification on asphalt binder dynamic and steady flow viscosities

Construction and Building Materials 71 (2014) 435–443

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Influence of polymer modification on asphalt binder dynamic and steady flow viscosities F. Cardone, G. Ferrotti, F. Frigio ⇑, F. Canestrari Department of Civil and Building Engineering and Architecture, Università Politecnica delle Marche, via Brecce Bianche, 60131 Ancona, Italy

h i g h l i g h t s  Oscillatory analysis and steady state viscosity measurements were performed on PMBs.  Polymer nature and content strongly influence PMBs rheological properties.  The applicability of the Cox–Merz rule for modified bitumens was evaluated.  Cross and Carreau models were suitable for low polymer modified bitumen.

a r t i c l e

i n f o

Article history: Received 19 May 2014 Received in revised form 31 July 2014 Accepted 23 August 2014

Keywords: Polymer modified bitumen Viscosity Rheological properties Cox–Merz relationship

a b s t r a c t Asphalt pavement performance such as rutting, crack initiation and propagation as well as fatigue behaviour are substantially affected by the rheological properties of the bitumen. In this sense, the use of polymer modification in road paving applications has been growing rapidly over the last decade as it allows significant enhancements in bitumen properties with consequent improvement in road service life. In fact, the use of polymer modified bitumens (PMBs) leads to pavements characterized by higher resistance to rutting and thermal cracking and lower fatigue damage, stripping and thermal susceptibility. This paper presents a laboratory investigation concerning the effect of polymer modification on the flow behaviour of bitumens. Two different polymers, an elastomer and a plastomer, were used as bitumen modifying agents at three different percentages (2%, 4% and 6% by bitumen weight). Oscillatory mechanical analysis as well as viscosity measurements under steady state conditions were performed taking into account different testing parameters such as temperature, loading frequency and shear rate. The results confirm that the rheological properties of PMBs are strongly influenced by polymer nature and polymer content. The bitumen viscosity on the dynamic domain was combined with that in the steady-state domain, confirming the applicability of the Cox–Merz relationship for the plain bitumen and the PMBs with low polymer content. Finally, the Cross and the Carreau models were found to be suitable to fit the steady state and the dynamic results in order to determine the viscosity function of the investigated bitumens. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction It has been well established that the rheological properties of the bitumen substantially affect the asphalt pavement performance [1]. Since bitumens for road paving applications experience a variety of thermo-mechanical states during their service life, it results extremely important to investigate their rheological properties under different temperature and loading conditions. Most pavement distresses, such as rutting at high temperatures and crack initiation and propagation at low temperatures, can be ⇑ Corresponding author. Tel.: +39 071 220 4507; fax: +39 071 220 4510. E-mail address: [email protected] (F. Frigio). http://dx.doi.org/10.1016/j.conbuildmat.2014.08.043 0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.

attributed not only to traffic loads but also to the thermal susceptibility of asphalt binders [2,3]. In order to improve asphalt mixture performance, the bitumen properties are often enhanced by means of polymer modification. Polymers are traditionally used to decrease the temperature susceptibility of the bitumen by increasing its stiffness at high service temperatures as well as reducing its stiffness at low service temperatures [4,5]. This leads to enhanced pavements having higher resistance to rutting and thermal cracking and lower fatigue damage, stripping and thermal susceptibility [6,7]. The polymers that are commonly used for bitumen modification can be divided into two main categories: elastomers and plastomers. Elastomers are characterized by high elastic response having

436

F. Cardone et al. / Construction and Building Materials 71 (2014) 435–443

the ability of resisting permanent deformations by stretching and recovering their initial shape [4], whereas plastomers contribute to form a tough, three-dimensional network to resist deformation [4] when blended with asphalt binder. The influence of polymer modification on bitumen properties can be evaluated through the investigation of rheological properties such as viscosity, complex modulus and phase angle that are highly influenced by the presence of a polymer network in the bituminous matrix. In particular, viscosity affects mixing, laying and compaction characteristics of asphalt mixtures as well as pavement performance and can be evaluated in steady state conditions by means of Brookfield viscosimeter or Dynamic Shear Rheometer (DSR). On the other hand, complex modulus and phase angle, that were measured by applying sinusoidal loads at various loading frequency and temperature ranges, allow a proper evaluation of the visco-elastic properties of the bitumens. The oscillatory measurements can be further employed to obtain the complex viscosity and the correlation between the dynamic and the steady state condition can be defined [8] through the Cox–Merz relationship [2,8,9]. The Cox–Merz relationship is empirical and establishes a correspondence between the steady-state viscosity at a specific shear rate (c_ ) and the magnitude of the complex viscosity at an angular frequency (x) equal to the considered shear rate, as shown in Eq. (1):

jg ðxÞj ¼ gðc_ Þjx¼c_

ð1Þ

This paper presents a laboratory investigation concerning the effect of polymer modification on the flow behaviour of a plain bitumen. Two different polymers were used as bitumen modifying agents at different percentages, in order to evaluate the effect of both polymer type and content on the rheological characteristics of asphalt binders. The effect of polymer modification was evaluated through oscillatory mechanical (i.e. dynamic) analysis and viscosity measurements under steady state conditions.

was removed from the can, poured into small containers, cooled to room temperature, cover with an aluminium foil and stored for not more than 7 days. Before testing, the stored material was heated once in order to produce testing specimens.

3. Test program and procedures Rheological properties of plain bitumen and polymer modified bitumens were studied according to the experimental program shown in Table 2. In a preliminary phase, complex modulus G and phase angle d were investigated with a Dynamic Shear Rheometer (DSR), performing frequency sweep tests over a range from 1 to 100 rad/s under isothermal conditions, in the temperature range from 4 to 82 °C with step of 6 °C. A plate–plate geometry was adopted with a plate diameter of 8 mm and a gap equal to 2 mm, from 4 to 34 °C and a plate diameter of 25 mm and a gap equal to 1 mm, from 34 to 82 °C. Frequency sweep tests were conducted in control strain within the linear viscoelastic range of the materials by applying a strain amplitude of 0.5%. In the second phase, the flow behaviour was evaluated in terms of dynamic viscosity measurements, in order to investigate the influence of polymer modification on both temperature susceptibility and shear rate dependency. In particular, bitumen viscosity g was measured in two different test configurations: (1) Coaxial cylinders with Brookfield device from 90 to 180 °C and shear rate of 10 s1. (2) Steady state with DSR, over a shear rate range from 0.001 to 1000 s1, at a temperature of 58 °C. Moreover, the complex viscosity g was determined from the frequency sweep tests performed over a wide range of temperatures (4–120 °C). Specially, it was calculated as the ratio between the norm of the complex modulus and the corresponding angular frequency, according to Eq. (2):

jg ðxÞj ¼ 2. Materials A 70/100 penetration-grade bitumen from an Italian oil refinery was selected as plain bitumen. Two polymers, an elastomer and a plastomer, were selected as modifying agents for the production of different modified bitumens (PMBs). A radial styrene–butadiene–styrene (SBS) polymer, which contains 30% styrene and has a density of 0.94 g/cm3, and a polyolefin (PO) polymer, having a density of 0.94 g/cm3, were added to the plain bitumen at three different percentages (2%, 4% and 6% by bitumen weight) with the aim to investigate the effect of polymer type as well as polymer content on the rheological properties of the plain bitumen. Table 1 shows the physical properties of the plain bitumen and the polymer modified bitumens. All modified bitumens were produced in the laboratory using a ROSS high-shear mixer, operating at a rotation speed of 3000 rpm at 180 °C. Initially, 700 g of bitumen contained in a 1000 ml cylindrical can were heated to fluid conditions. Up on reaching 180 °C, each type of polymer was added slowly to the bitumen in order to prevent any polymer aggregation during the mixing process. Subsequently, dicumyl peroxide in granules was added as cross-linker dosed at 0.3% by polymer weight. Mixing was then continued at 180 °C for 3 h. After mixing, the sample

Table 1 Physical properties of the plain bitumen and the polymer modified bitumens. Materials

Penetration @25 °C 0.1 mm (EN 1426)

Softening point °C (EN 1427)

Dynamic viscosity @135 °C Pas (EN 12595)

Plain PMB_SBS2 PMB_SBS4 PMB_SBS6 PMB_PO2 PMB_PO4 PMB_PO6

72 58 54 44 64 53 41

47.7 50.4 69.5 94.6 52.4 55.8 96.6

0.31 0.55 0.91 1.40 0.54 0.78 1.00

jG j

x

ð2Þ

Different devices and test configurations were used to investigate viscosity with the objective of extending the range of tested temperatures and loading frequencies as well as comparing the viscosity measured under rotational and oscillatory conditions. In fact, the rheological properties evaluated under dynamic conditions can be related to those measured under steady state conditions within specific ranges of shear rate and frequency, according to the Cox–Merz relationship (Eq. (1)). 4. Results and analysis 4.1. Preliminary dynamic characterization The isochronal plots of complex shear modulus G vs. temperature at a loading frequency of 1 rad/s are shown in Fig. 1 for all the studied materials. As expected, as the amount of polymer content increases, modified bitumens show an increase in G at all temperatures and a slight decrease in temperature susceptibility for temperatures equal or higher than 40 °C with respect to the plain bitumen, mainly for the highest polymer content. In particular, PMB_PO6 shows a sharp decrease in the slope of the complex modulus isochronal at high temperatures with the establishment of a plateau region which is indicative of a dominant polymer network [4,10]. Moreover, at low temperatures the effect of polymer results not very evident regardless of polymer type and content because of the high stiffness of the plain bitumen that does not allow the polymer network to influence noticeably the rheological properties of the bitumen [4].

437

F. Cardone et al. / Construction and Building Materials 71 (2014) 435–443 Table 2 Experimental program. Characteristics 

G,d Viscosity g Viscosity g Viscosity g

Device (–)

Type of test (–)

Temperature (°C)

Shear rate (s1)

Angular frequency (rad/s)

DSR Brookfield DSR DSR

Oscillation Rotation Rotation Oscillation

4–82 90–180 58 4–120

– 10 0.001–1000 –

1–100 – – 10

1.E+05

1.E+05 Plain

Plain 1.E+04

PMB_SBS2 PMB_SBS4

1.E+03

Complex modulus [kPa]

Complex modulus [kPa]

1.E+04

PMB_SBS6

1.E+02 1.E+01 1.E+00 1.E-01

PMB_PO2 PMB_PO4

1.E+03

PMB_PO6 1.E+02 1.E+01 1.E+00 1.E-01

1.E-02

1.E-02 0

10

20

30

40

50

60

70

80

90

0

10

20

30

Temperature [°C]

40

50

60

70

80

90

Temperature [°C]

Fig. 1. Complex modulus vs. temperature at 1 rad/s for SBS (left) and PO (right) modified bitumens.

polymer dissolved in the bitumen forms a continuous rubber-elastic network that allows an enhancement of elastic properties of the material. A similar behaviour is also shown by PO modified bitumens (Fig. 2-right) with lower polymer contents (2% and 4%), whereas PMB_PO6 provides a sharply decrease of the phase angle, for temperatures higher than 40 °C. This could be attributed to a more structured polymer phase due to the plastomer characteristics. In fact, the polymeric phase in PMBs can be continuous or dispersed, depending on the polymer content and its ability to swell with the maltene molecules of the bitumen [2,12]. The effect of polymer modification on bitumen properties is related to the characteristics of such polymeric phase. These results confirm that the rheological properties of polymer modified bitumens are strongly influenced by polymer nature, polymer content and bitumen composition [10]. 4.2. Viscosity properties The viscosity properties of the studied bitumens were evaluated in order to investigate the viscosity dependence on testing

100

100

90

90

80

80

70

70

Phase angle [°]

Phase angle [°]

The isochronal plots of phase angle vs. temperature at 1 rad/s are shown in Fig. 2 for all the studied materials. The phase angle of the plain bitumen tends to increase with temperature, approaching a viscous behaviour (d = 90 °C) at high temperatures. On the other hand, all the studied PMBs demonstrate lower phase angle values than the plain bitumen on the overall temperature range, confirming that the presence of a polymer network tends to increase the elastic properties of the material [4,10]. In accordance with other authors [4,11], such effect results more evident at high temperatures since the low viscosity of the plain bitumen allows the elastic network of the polymers to influence the mechanical properties of the modified bitumens. In particular, at low polymer content (PMB_SBS2 and PMB_PO2), PMBs curves have a trend similar to the plain bitumen (increase of the viscous properties with temperature) but with a lower phase angle at each temperature. On the contrary, higher polymer contents lead to different behaviours. In fact, at intermediate and high temperatures, the phase angle of the SBS polymer modified bitumens at 4% and 6% does not increase with temperature but it tends to decrease. This result suggests that for such SBS contents, the

60 50 40 Plain 30

PMB_SBS2

60 50 40

Plain

30

PMB_PO2

20

PMB_SBS4

20

PMB_PO4

10

PMB_SBS6

10

PMB_PO6

0

0 0

10

20

30

40

50

Temperature [°C]

60

70

80

90

0

10

20

30

40

50

Temperature [°C]

Fig. 2. Phase angle vs. temperature at 1 rad/s for SBS (left) and PO (right) modified bitumens.

60

70

80

90

438

F. Cardone et al. / Construction and Building Materials 71 (2014) 435–443

temperature and shear rate and define the effect of polymer modification on the bitumen behaviour. 4.2.1. Temperature susceptibility The viscosity values of the plain bitumen and the polymer modified bitumens at temperatures of 58 °C and 135 °C are shown in Fig. 3. These temperatures were chosen because they are representative of fundamental Hot Mix Asphalt (HMA) pavement conditions. In particular, it is well known that the bitumen viscosity at temperatures around 60 °C is a key parameter for the evaluation of the performance of HMA pavements during summer as a low viscosity can induce problems such as rutting and/or flushing [1]. On the other hand, the viscosity at 135 °C is close to that required for the mixing and the compaction of asphalt concrete mixtures. The viscosity at 58 °C was derived from the oscillatory measurements using the Cox–Merz relationship while the viscosity at 135 °C was directly measured with the Brookfield viscometer. Data show that the effect of polymers results in a bitumen viscosity increase as the polymer content increases, at both temperatures. At high temperatures (about 135 °C), SBS polymer provides a more pronounced increase in the bitumen viscosity with respect to polyolefin, which on the contrary offers a more evident viscosity increase at 58 °C in correspondence of each polymer content. These results can be attributed to the lower consistency of polyolefin at higher temperatures. In order to evaluate the temperature susceptibility of bitumen viscosity in a wide range of temperatures, both Brookfield viscosity (from 90 °C to 180 °C) and viscosity derived from oscillatory measurements (from 4 °C to 120 °C) were considered for all the tested materials, at a shear rate of 10 s1. Fig. 4 shows a good alignment between the two series of viscosity data for the plain bitumen (Fig. 4a) and for the PMBs with low polymer content (Fig. 4b and c), allowing the applicability of the Cox–Merz relationship for these materials. On the contrary, as the polymer content increases (Fig. 4d, f and g), the experimental data seem to diverge from a single viscosity trend, denoting a poor correlation between rotational and oscillatory results. The viscosity data ranging from 4 to 120 °C were used to determine the temperature susceptibility of the bitumens by means of the ASTM model given in Eq. (3):

log log g ¼ A þ VTS  log T R

ð3Þ

where g is the viscosity (cP), A is the regression intercept, VTS is the regression slope (viscosity–temperature susceptibility parameter) and TR is the temperature (°R). Fig. 5 shows that all the polymer modified bitumens provide lower regression slope values VTS than the plain bitumen meaning that the viscosity–temperature susceptibility tends to decrease as the polymer content increases. As expected, the presence of a

polymer network in the bitumen leads to a lower decrease in viscosity with temperature, resulting in higher aptitude to resist to the accumulation of permanent deformation. In particular, high content of PO polymer (6%) is needed to observe a valuable temperature susceptibility decrease, whereas the addition of low amounts of SBS polymer is enough to significantly affect the bitumen temperature susceptibility. Finally, as shown in Fig. 6 (trend of dynamic viscosity vs phase angle), the change in bitumen viscosity with temperature can be associated to the different rheological states (from visco-elastic to purely viscous) that characterize the materials throughout the test [13]. In particular, the plain bitumen shows a gradual decrease in viscosity with the increase in the phase angle and, for high temperatures, the curve approaches the vertical asymptote for d equal to 90°. At low polymer contents, PMBs (PMB_SBS2, PMB_PO2 and PMB_PO4) behave similarly to the plain bitumen as they are characterized by a predominant viscous behavior (i.e. high phase angle) for high temperatures. Whereas, as the amount of polymer increases the rheological behavior moves to a viscoelastic character (i.e. lower phase angle). In particular, PO polymer modified bitumen at 6% shows a significant change in behavior and tends to an elasticity-dominant region (i.e. phase angle below 45°). 4.2.2. Shear rate influence It is well known that unmodified bitumens behave as classic Newtonian fluids at temperatures higher than 60 °C, whereas polymer modified bitumens exhibit a shear rate dependent viscosity that decreases with increasing shear rate [14]. The dependence of the dynamic viscosity on the shear rate at a temperature of 58 °C is given in Fig. 7. For the plain bitumen and the PMBs with low polymer contents (2% for SBS and 2–4% for PO), the dynamic viscosity at 58 °C can be considered almost independent on the shear rate. As the amount of polymer increases, the bitumens start to show a shear thinning behaviour, according to Lu and Isacsson [13]. In particular, in the studied range of shear rate, the bitumen modified with PO at high content (PMB_PO6) demonstrates a more marked shear thinning behaviour (the viscosity roughly decreases as the shear rate increases) compared to the bitumen modified with SBS. In general, for unmodified bitumens the shear rate dependency of the viscosity includes a constant viscosity plateau at very low shear rates (zero shear viscosity) and a lower constant viscosity plateau at high shear rate (infinite shear viscosity) [15]. With the purpose of studying the viscosity function, the viscosity under steady state flow condition and the norm of the complex viscosity were plotted in one graph (Fig. 8) allowing also the investigation of the validity of the Cox–Merz relationship for unmodified and modified asphalt binders in the domain where the shear rate and the reduced frequencies overlap [16].

2.0

PMB_SBS

2000

T=58°C

Viscosity @ 10 s-1 [Pa*s]

Viscosity @ 10 rad/s [Pa*s]

2500

PMB_PO 1500

1000

500

0

T=135°C

PMB_SBS 1.5

PMB_PO

1.0

0.5

0.0 0

2

4

Polymer content [%]

6

0

2

4

Polymer content [%]

Fig. 3. Effect of polymers content on viscosity at 58 °C (left) and 135 °C (right).

6

439

F. Cardone et al. / Construction and Building Materials 71 (2014) 435–443

(a)

1.E+08

Plain

Viscosity [Pa*s]

1.E+06

1.E+04

1.E+02

1.E+00

oscillatory Brookfield

1.E-02 0

20

40

60

80

100

120

140

160

180

200

Temperature [°C]

(b) 1.E+08

(c) 1.E+08 PMB_SBS2

1.E+04

1.E+02

oscillatory

1.E+00

PMB_PO2

1.E+06

Viscosity [Pa*s]

Viscosity [Pa*s]

1.E+06

1.E+04

1.E+02

oscillatory

1.E+00

Brookfield

Brookfield

1.E-02

1.E-02 0

20

40

60

80

100

120

140

160

180

200

0

20

40

60

Temperature [°C]

(d) 1.E+08 PMB_SBS4

Viscosity [Pa*s]

Viscosity [Pa*s]

120

140

160

180

200

PMB_PO4

1.E+06

1.E+04

1.E+02

oscillatory

1.E+00

1.E+04

1.E+02

oscillatory

1.E+00

Brookfield

Brookfield

1.E-02

1.E-02 0

20

40

60

80

100

120

140

160

180

200

0

20

40

60

80

100

120

140

160

180

200

Temperature [°C]

Temperature [°C]

(g) 1.E+08

1.E+08

PMB_SBS6

1.E+04

1.E+02

oscillatory

1.E+00

PMB_PO6

1.E+06

Viscosity [Pa*s]

1.E+06

Viscosity [Pa*s]

100

(e) 1.E+08

1.E+06

(f)

80

Temperature [°C]

1.E+04

1.E+02

1.E+00

oscillatory

Brookfield

Brookfield

1.E-02

1.E-02 0

20

40

60

80

100

120

140

160

180

200

0

20

40

Temperature [°C]

60

80

100

120

140

160

180

200

Temperature [°C]

Fig. 4. Viscosity vs. temperature at shear rate of 10 s1.

Steady state viscosity by means of DSR was obtained as the ratio between the measured shear stress and the applied shear rate in a rotational test, taking into account only the viscosity data obtained when the specimen was undamaged. An example of the shear stress vs shear rate curves is reported in Fig. 9. Such curves were used to define the range of valid data: the occurrence of the damage in the specimens can be easily identified with the point at which the shear stress starts to decrease with the increase of the shear rate. As far as the complex viscosity is concerned, the

time–temperature superposition principle was used to shift |g| data (obtained applying Eq. (2) to G results) at a reference temperature of 58 °C. Then, the complex viscosity in the dynamic space was converted into viscosity in the steady-state space by using the Cox–Merz relationship, according to Eq. (1). Fig. 8 shows that none of the studied bitumens approaches the infinite viscosity g1 at the tested shear rates. However, it is interesting to notice that the plain bitumen and both the PMBs with low polymer content (PMB_SBS2 and PMB_PO2) exhibit a clear

440

F. Cardone et al. / Construction and Building Materials 71 (2014) 435–443 -3.6

12.0

Regressione slope VTS [-]

Regressione parameter A [-]

-3.4 10.0

8.0

6.0

4.0

2.0

-3.2 -3.0 -2.8 -2.6 -2.4 -2.2 -2.0

0.0 Plain

Plain

PMB_SBS2PMB_SBS4PMB_SBS6 PMB_PO2 PMB_PO4 PMB_PO6

PMB_SBS2PMB_SBS4PMB_SBS6 PMB_PO2 PMB_PO4 PMB_PO6

Fig. 5. Regression parameters of the viscosity–temperature relationship.

1.E+07

1.E+07

Plain

1.E+06

Plain

1.E+06

PMS_SBS2

PMS_PO2 1.E+05

PMS_SBS4

Viscosity [Pa*s]

Viscosity [Pa*s]

1.E+05

PMS_SBS6

1.E+04 1.E+03

PMS_PO6

1.E+03

1.E+02

1.E+02

1.E+01

1.E+01

1.E+00

PMS_PO4

1.E+04

1.E+00 0

10

20

30

40

50

60

70

80

90

100

0

10

20

Phase angle [°]

30

40

50

60

70

80

90

100

Phase angle [°]

Fig. 6. Viscosity as a function of phase angle at 10 rad/s for SBS (left) and PO (right) modified bitumens.

1.E+05

1.E+05 Plain

Plain

T=58°C

PMB_SBS2

T=58°C

PMB_PO2 PMB_PO4

PMB_SBS6

1.E+04

Viscosity [Pa*s]

Viscosity [Pa*s]

PMB_SBS4

1.E+03

1.E+02 1.E+00

1.E+01

1.E+02

Shear rate [1/s]

PMB_PO6

1.E+04

1.E+03

1.E+02 1.E+00

1.E+01

1.E+02

Shear rate [1/s]

Fig. 7. Viscosity at 58 °C as a function of shear rate for SBS (left) and PO (right) modified bitumens.

Newtonian behaviour at low shear rate, showing a plateau in correspondence of the zero shear viscosity value g. Nevertheless, for shear rate higher that 10 s1, such materials (Fig. 8a, b and e) show a shear thinning behaviour, characterized by a decrease in viscosity with increase in shear rate. On the other hand, the modified bitumens with higher polymer contents (Fig. 8c, d and f) exhibit a shear rate dependent viscosity for all the shear rates tested. In particular, these materials show a shear thinning behaviour without experiencing Newtonian behaviour in the shear rate range investigated. This is consistent with previous studies that subdivided modified binders in two categories based on rheological criteria [17].

Moreover, from the results shown in Fig. 8, it is possible to affirm that the correlation between the steady state viscosity and the viscosity in oscillation is good for the plain bitumen and the modified bitumens with low polymer content (2%) as confirmed by the satisfactory overlapping between the steady state and the dynamic data at low shear rates. On the contrary, as the amount of polymer modification increases, a poor correlation between the two viscosity trends occurs due to an increased degree of non-Newtonian behavior, according to Lu and Isacsson [13] results. Finally, the materials that exhibit a zero shear viscosity plateau (plain bitumen, PMB_SBS2 and PMB_PO2) were employed in

441

F. Cardone et al. / Construction and Building Materials 71 (2014) 435–443

(a)

(b)

1.E+04

1.E+04

PMB_SBS2

1.E+03

Viscosity [Pa*s]

Viscosity [Pa*s]

Plain

1.E+02

1.E+03

1.E+02

steady state

steady state

oscillatory

oscillatory

1.E+01 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05

1.E+01 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05

Shear rate [1/s]

(c)

Shear rate [1/s]

1.E+04

(d) 1.E+04

1.E+03

PMB_SBS6 Viscosity [Pa*s]

Viscosity [Pa*s]

PMB_SBS4

1.E+02

1.E+03

1.E+02

steady state

steady state

oscillatory

oscillatory

1.E+01 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05

1.E+01 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05

Shear rate [1/s]

(e)

Shear rate [1/s]

1.E+04

(f)

1.E+04

PMB_PO4 Viscosity [Pa*s]

Viscosity [Pa*s]

PMB_PO2 1.E+03

1.E+02

1.E+03

1.E+02

steady state

steady state

oscillatory

oscillatory

1.E+01 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05

1.E+01 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05

Shear rate [1/s]

Shear rate [1/s]

Fig. 8. Comparison between steady state viscosity and viscosity derived from oscillatory measurements at 58 °C.

obtaining the viscosity functions, by fitting the steady state and the dynamic results with two of the most widespread models. In particular, the Cross model [18] and the Carreau model [19], shown in Eqs. (4) and (5) respectively, were used:

10000 T=58°C

Shear stress [Pa]

1000

g ¼ g1 þ

100

g ¼ g1 þ

10

1

0.1 0.001

0.01

0.1

1

10

100

Shear rate [1/s] Fig. 9. Typical curve of shear stress vs shear rate.

1000

g0  g1 m ½1 þ ðkc_ Þ  g0  g1 2 b=2

ð4Þ

ð5Þ

½1 þ ðac_ Þ 

where g0 is the zero shear viscosity [Pas], g1 is the infinite viscosity [Pas], k, m, a and b are regression coefficients. The results in terms of zero shear viscosity g0 and R-square for both the analyzed models are shown in Table 3, where it is evident that both the Cross model and the Carreau model (and its derivation) are suitable for studying the bitumens considered in

442

F. Cardone et al. / Construction and Building Materials 71 (2014) 435–443

Table 3 Zero shear viscosity and R-square of the regression results for Cross and Carreau models. Materials Plain PMB_SBS2 PMB_PO2

(Pas)

R2 (–)

242 241 565 547 350 345

0.946 0.946 0.979 0.989 0.876 0.895

g0

Model Cross Carreau Cross Carreau Cross Carreau

8.E+02

Plain_Cross Plain_Carreau PMB_SBS2 Cross

[Pa*s]

6.E+02

PMB_PO2 Cross PMB_PO2 Carreau

0

Estimated

PMB_SBS2 Carreau

4.E+02

2.E+02

0.E+00 0.E+00

2.E+02

4.E+02

Measured

6.E+02

8.E+02

0 [Pa*s]

– The addition of SBS polymer results in a more pronounced increase in bitumen viscosity at high temperatures (135 °C) with respect to polyolefin polymer, which on the contrary offers a more evident viscosity increase at intermediate temperatures (58 °C), thanks to the lower consistency of the polyolefin at higher temperatures. – All the investigated PMBs are characterized by a less temperature dependent viscosity that results in higher aptitude to resist to the accumulation of permanent deformation. In particular, high content of PO polymer (6%) is needed to observe a valuable temperature susceptibility decrease, whereas the addition of low amounts of SBS polymer is enough to significantly affect the bitumen temperature susceptibility. – The dynamic viscosity at 58 °C can be considered almost independent on the shear rate for the plain bitumen and the PMBs with low polymer contents (2% for SBS and 2– 4% for PO); such materials demonstrate also a satisfactory correlation between the steady state viscosity and the viscosity derived from oscillatory measurements, confirming the applicability of the Cox–Merz relationship. On the contrary, modified bitumens with higher polymer contents exhibit a shear thinning behavior for all shear rates tested and poor correlation between the two sets of viscosity data, due to an increased degree of non-Newtonian behavior. – The plain bitumen and the PMBs with low polymer contents exhibit a shear viscosity plateau in the low shear rate region and both the Cross model and the Carreau model are found to be suitable for the evaluation of the viscosity function.

Fig. 10. Comparison between the estimated (Cross & Carreau models) and the measured zero shear viscosity.

this investigation, as verified also in other studies [20]. In fact, both models predict a similar zero shear viscosity value for each bitumen and all the R-square values are higher than 0.87. A visual comparison between the zero shear viscosity estimated applying the two models and the viscosity measured at very low shear rates is given in Fig. 10. The data result very close to the equality line for all the analyzed materials, confirming that both models are able to predict the zero shear viscosity for the plain bitumen and the PMBs with low polymer content. 5. Conclusions A radial styrene–butadiene–styrene (SBS) and a polyolefin (PO) polymer were selected as modifying agents for the laboratory production of six modified bitumens (PMBs) with the aim of investigating the effect of polymer type as well as polymer content on the rheological properties of a plain bitumen. Oscillatory measurements were performed by means of Dynamic Shear Rheometer (DSR) and viscosity measurements were carried out in steady state conditions by means of both Brookfield viscometer and DSR over a wide range of temperatures and loading conditions. Based on the experimental results, several conclusions can be drawn: – The presence of a polymer network in the bitumen leads to increase the stiffness (complex modulus) and the elastic properties of the material (lower phase angle) as well as to decrease its temperature susceptibility. In particular, the presence of SBS results in a gradual change in bitumen characteristics as the amount of polymer increases whereas high content of PO (6%) is required for a significant improvement in bitumen performance, highlighting the significant role of the polymer content and type.

Acknowledgements This work was supported by the MIUR project ‘‘Damage and healing of innovative nano-structured and polymer-modified bituminous materials’’ (Grant RBFR10JOWO) under the ‘‘FIRB – Futuro in Ricerca 2010’’ funding program.

References [1] Brown ER, Kandhal PS, Roberts FL, Kim YR, Lee DY, Kennedy TW. Hot mix asphalt materials, mixture design, and construction. 3rd ed. National Center for Asphalt Technology; 2009. [2] Pérez-Lepe A, Martínez-Boza FJ, Gallegos C, González O, Muñoz ME, Santamaría A. Influence of the processing conditions on the rheological behaviour of the polymer-modified bitumen. Fuel 2003;82:1339–48. [3] Ait-Kadi A, Brahimi H, Bousmina M. Polymer blends for enhanced asphalt binders. Polym Eng Sci 1996;36:1724–33. [4] Airey GD. Rheological evaluation of ethylene vinyl acetate polymer modified bitumens. Constr Build Mater 2002;16:473–87. [5] Collins JH, Bouldin MG, Gelles R, Berker A. Improved performance of paving asphalts by polymer modification. AAPT 1991;60:43–79. [6] Yildrim Y. Polymer modified asphalt binders. Constr Build Mater 2007;21:66–72. [7] Santagata E, Baglieri O, Dalmazzo D. Tsantilis LEvaluation of the anti-rutting potential of polymer-modified binders by means of creep-recovery shear tests. Mater Struct 2013;46:1673–82. [8] Shan L, Tan Y, Kim YR. Applicability of the Cox–Merz relationship for asphalt binder. Constr Build Mater 2012;37:716–22. [9] Cox WP, Merz EH. Correlation of dynamic and steady flow viscosities. J Polym Sci 1958;28:619–22. [10] Airey GD. Rheological properties of styrene butadiene styrene polymer modified road bitumens. Fuel 2003;85:1709–19. [11] Goodrich JL. Asphaltic binder rheology, asphalt concrete rheology and asphalt concrete mix properties. J Assoc Asphalt Pavement Technol 1991;60: 80–120. [12] Adedeji A, Grunfelder T, Bates FS, Macosko CW, Stroup-Gardiner M, Newcomb DE. Asphalt modified by SBS triblock copolymer: structures and properties. Polym Eng Sci 1996;36:1707–23. [13] Lu X, Isacsson U. Influence of styrene–butadiene–styrene polymer modification on bitumen viscosity. Fuel 1997;76:1353–9.

F. Cardone et al. / Construction and Building Materials 71 (2014) 435–443 [14] Wasiuddin NM, Saha R, King Jr W, Mohammad L. Effects of temperature and Ò shear rate on viscosity of Sasobit – modified asphalt binders. Int J Pavement Res Technol 2012;5(6):369–78. [15] Mezger TG. The rheology handbook, 3rd ed. Vincentz Network; 2011. [16] Stastna J, Zanzotto L, Vacin OJ. Viscosity function in polymer – modified asphalts. J Colloid Interface Sci 2003;259:200–7. [17] Merusi F, Giuliani F. Intrinsic resistance to non-reversible deformation in modified asphalt binders and its relation with specification criteria. Constr Build Mater 2011;25:3356–66.

443

[18] Cross MM. Rheology of non-Newtonian fluids: a new flow equation for pseudoplastic systems. J Colloid Interface Sci 1965;20:417–37. [19] Carreau PJ, MacDonald IF, Byron Bird R. A nonlinear viscoelastic model for polymer solutions and melts-II. Cham Eng Sci 1968;23(8):901–11. [20] Giuliani F, Merusi F. Flow characteristics and viscosity functions in asphalt binders modified by wax. Int J Pavement Res Technol 2009;2(2):51–60.