chemomechanical approaches to predict asphalt pavement

Chemomechanical Approaches to Predict Asphalt Pavement Behavior and

Performance

Troy Pauli, Western Research Institute Linbing Wang and Wenjuan Sun,

Virginia Polytechnic Institute

“ARC Deliverables/Products Presentation and Workshop”TRB 2015, Thursday, January 15, 2015, 9:00am - 5:00pm,

Marriott Marquis, Washington DC.

This research has been supported by

Federal Highway Administration U. S. Department of Transportation Under Cooperative AgreementNo. DTFH61-07-H-00009

Delft University of TechnologyFaculty of Civil Engineering and Geosciences Section of Road and Railway Engineering

& The Delft Healing Consortium

Acknowledgements

2

Asphalt Microstructure Model Team

• Asphalt composition (chemistry) drives binder rheological behavior, which in turn drives mechanical (performance) behavior.

• Crude source dictates asphalt composition (chemistry) but may be modified to produce desired rheological and mechanical (performance) behavior.

• Crude source variation hinders prediction of pavement behavior.• Fundamental theories of material properties coupled with multi-scale

modeling efforts implemented now insures that the technology needed to effectively modify and model asphalt pavement behavior will be available in the future.

Asphalt Microstructure Model Team

• Multiscale modeling endeavors provide a link between fundamental material properties and mechanical (performance) properties.

• A chemomechanical theory of asphalt binder behavior has developed which integrates asphalt composition (chemistry) to physicochemistry and binder rheology, which relates to pavement mechanical behavior.

• A deliverable of this research is source code of phase-behavior of asphalt binder (i.e., asphalt composition (chemistry) ) which is compatible with finite element methods to predict mechanical response.

• A chemomechanical modeling platform will provide chemical, physical and rheological input parameters for finite element models based on fundamental material properties and theories.

Asphalt Microstructure Model Team

Physicochemical Model of RAP Blending

2

Prediction of RAP Asphalt Flow

10.00

11.00

12.00

13.00

14.00

15.00

16.00

17.00

18.00

19.00

20.00

0.00 0.20 0.40 0.60 0.80 1.00RAP Mass Fraction, χRAP

Pred

icte

d ln

( η)

AAB-1-waxWMA/AAA-1-(RAP-1)

AAB-1-waxWMA/AAD-1-(RAP-2)

AAG-1-waxWMA/AAD-1-(RAP-2)

YNP-waxWMA/AAD-1-(RAP-2)

( ) ( )( ) ( ) ( )( )( ) 2.5

0 0 1blend RAP neat blend RAP a neat aRAP neat RAP neatx x K x x x xη η η

−= + − +

RAP-Asphalt Flow Properties

Asphaltene Mass Fraction (Iso-Octane)

0.0 0.1 0.2 0.3 0.4 0.5 0.6

G*

at 6

0C a

t 10

rad/

s., P

a

1e+0

1e+1

1e+2

1e+3

1e+4

1e+5

1e+6

1e+7

Y=75.147*exp(18.975X), R2=0.8596

Prediction of RAP Asphalt Flow

60.00

65.00

70.00

75.00

80.00

85.00

90.00

0.00 0.20 0.40 0.60 0.80 1.00

RAP Mass Fraction, χRAP

Pred

icte

d Ph

ase

Angl

e, δ

AAB-1-waxWMA/AAA-1-(RAP-1)

AAB-1-waxWMA/AAD-1-(RAP-2)

AAG-1-waxWMA/AAD-1-(RAP-2)

YNP-waxWMA/AAD-1-(RAP-2)

( )( )0

0

1

1SECI

asphaltene

χ

χ

≈ −

≈ −

δ δ

δ

RAP-Asphalt Flow Properties

Asphaltene Mass Fraction (Iso-Octane)

0.0 0.1 0.2 0.3 0.4 0.5 0.6

Phas

e An

gle

at 6

0C, 1

0 ra

d/s.

60

65

70

75

80

85

90

Y=-288.05X2+36.892X+87.192, R2=0.9067

Physicochemical Model of Wax Surface Structuring

2

Paraffin surface freezing “detected” byX-ray reflectivity and grazing incidence diffraction

X.Z. Wu, B.M. Ocko, E.B. Sirota, S.K. Sinha, M. Deutsch, B.H. Cao, M.W. Kim, Surface Tension Measurements of Surface Freezing in Liquid Normal Alkanes, Science, 261 (1993a) 1018-1020.

Wax Surface-Freezing in Complex Media

Artificial Bees: Paraffin in paraffin oil

Wax Surface-Freezing in Complex Media

Wax Surface-Freezing in Complex Media

1%

5%

10%

20%

Wax Surface-Freezing in Complex Media

Wax Surface-Freezing in Complex Media

y = 0.5595x - 0.0142R² = 0.8947

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 2 4 6

Cry

tallin

e M

ater

ial v

ia D

SC

Γ, fitting coefficient

y = 1.1936x + 0.6833R² = 0.9711

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

0 2 4 6

Per

cent

Wax

Ext

ract

ed fr

om IE

C N

eutra

ls

Γ, fitting coefficient

Wax Surface-Freezing in Complex Media

Finite Element Model of Fatigue-Healing

Driven by Wax Structuring

2

Finite Element Modeling

2036

E∞

η

σσEm

E∞

Em η

σσ D

{ }σ εd d di

ji ij plastic d

TS X J

a

=

= +å&

& &G

Finite Element Modeling

1 2 3

2139

0

2

4

6

8

10

12

0 1 2 3 4

Dis

plac

emen

t/loa

ding

rate

slo

pe X

100

0

Roundness

“5”

Finite Element Modeling

2240

0

0.1

0.2

0.3

0.4

0.5

0.6

0 5 10 15 20 25 30

Hea

ling

Rat

e In

dice

s

Mass % Wax per IEC Neutral

dh/dt

dh2/dt

Healing and Presence of Wax

• The oils phase viscosity is driven by molecular weigh.• The maltene (i.e., the oils plus resins) may be considered a polar

associated solution, BUT, apparent molecular weight may be a big player in viscosity.

• The asphalt (maltene plus asphaltenes) is conveniently modeled as a nano-particle suspension. AGAIN, apparent molecular weight considerations may play a significant role in viscous flow.

• Activation energy models of viscosity work well to explain viscosity-temperature relationships of the nano-particle suspension model considered.

• The Oil’s molecular weight correlates with thermal contraction properties of asphalt, thus, thermal fatigue correlates with thermal contraction, ergo, oils molecular weight correlates with , thermal fatigue.

• Asphaltene content strongly contributes to stiffness of RAP mixtures.

Findings-Conclusions

Findings-Conclusions

• The “LIQUID” (behavior) of asphalt relates to fatigue-healing.• Surface crystallization (Surface Freezing) of paraffin wax in complex media

involves diffusion controlled crystallization. Bees are observed in both thermally and solvent treated materials.

• Crystals develop at a liquid-vapor interface due to lowering of the liquid-vapor surface free energy state relative to a liquid-solid + solid-vapor state.

• Oils and wax concentration are mutually related in terms of molecular weight, both of which relate to healing.

• Force-flux Coupling makes it difficult to pin-point a single compositional contributor to binder mechanical (rheological) behavior.

• IN SHROT: a chemomechanics model of bitumen behavior based in non-equilibrium statistical thermodynamics of asphalt microstructure is proposed to predict damage-healing phenomena.