NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000...
Transcript of NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000...
![Page 1: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/1.jpg)
NANO MICRO MESO MACRO
PG grading of asphalt binders
??? Beam fatigue / Cyclic load in indirect tension
Current Specifications for fatigue cracking:
Properties of materials used
Properties of mastic Properties of mixture
Important interactions between asphalt binder and aggregate fines related to fatigue cracking cannot be ignored !
1) stiffening of the matrix due to addition of fines
2) crack pinning and crack growth
![Page 2: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/2.jpg)
A variety of approaches to characterize fatigue cracking in asphalt mixes are possible depending on the type of material, test and analysis:
Test Method
• Bending beam test (asphalt mixes)
• Repeated load indirect tension test (asphalt mixes)
• DMA / DSR (asphalt mastics)
Test Mode
• Controlled stress
• Controlled strain
Analysis Method
• Direct methods
• Energy methods
• Crack growth modeling
• Rest periods (recovery)
![Page 3: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/3.jpg)
Test Method
• Bending beam test (asphalt mixes)
• Repeated load indirect tension test (asphalt mixes)
• DMA / DSR (asphalt mastics)
Test Mode
• Controlled stress
• Controlled strain
Analysis Method
• Direct methods
• Energy methods
• Crack growth modeling
• Rest periods (recovery)
![Page 4: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/4.jpg)
CoarseAggregates
FineAggregates
Fillers
Scale IAsphalt Concrete
Scale IIAsphalt Matrix
Scale IIIBinder or Mastic
Scale II represents aggregate < No. 16Sieve and mastic
![Page 5: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/5.jpg)
Test Method: DMA / DSR (Asphalt Mastics)
Sample preparation:
• Similar to HMA compaction using SGC
• Up to 30 samples can be cored from a single SGC compacted specimen
![Page 6: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/6.jpg)
Test Method : DMA / DSR (Asphalt Mastics)
Test procedure and setup:
• Test is conducted by applying constant strain or constant stress and recording response
• G* and phase angle are determined in real time during the test
1.E+06
1.E+07
1.E+08
1.E+09
1.E-04 1.E-01 1.E+02 1.E+05
Reduced frequency (radian)
LVE
dyna
mic
mod
ulus
(Pa)
0
10
20
30
40
50
60
LVE
phas
e an
gle
(deg
ree)
dynamic shear modulusphase angle
slope = m-value
-100000
-75000
-50000
-25000
0
25000
50000
75000
100000
490.85 490.95 491.05 491.15 491.25 491.35 491.45
Time (sec)Sh
ear s
tres
s (P
a)-1.00
-0.75
-0.50
-0.25
0.00
0.25
0.50
0.75
1.00
Shea
r str
ain
(%)
shear stressshear strain
phase lag
![Page 7: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/7.jpg)
Test Method
• Bending beam test (asphalt mixes)
• Repeated load indirect tension test (asphalt mixes)
• DMA / DSR (asphalt mastics)
Test Mode
• Controlled stress
• Controlled strain
Analysis Method
• Direct methods
• Energy Methods
• Crack growth modeling
• Rest periods (recovery)
![Page 8: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/8.jpg)
Analysis Methods : Direct Methods• Direct method to estimate fatigue life is based on the
number of load cycles to reach a pre-specified failure criteria
• Classical criteria include 50% reduction in stiffness, FIP (first inflection point), Nt (transition point), and SIP (second inflection point.
0.0E+00
2.0E+07
4.0E+07
6.0E+07
8.0E+07
1.0E+08
0 2000 4000 6000 8000 10000
No. of loading cycles
Non
linea
r dyn
amic
mod
ulus
(Pa)
20
25
30
35
40
45
50
55
60
Non
linea
r pha
se a
ngle
(deg
ree)
Nonlinear dynamic modulusNonlinear phase angle
peak phase angle
transition point
50% Stiffness Reductionmicrocracking
macrocracking
![Page 9: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/9.jpg)
Analysis Methods : Direct Methods
• The transition point, Nt, is calculated as the point where the value N(G*N/G*1) vs. N reaches a maxium
0
200
400
600
800
1000
1200
1400
1600
1800
2000
0 1000 2000 3000 4000 5000 6000 7000
Cycles
NxG
'/G
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
G'/G
Fatigue Life
G'/G at Failure
Fatigue Life at Maximum NxG'/G
![Page 10: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/10.jpg)
1.E+03
1.E+04
1.E+05
1.E+06
1.E+03 1.E+04 1.E+05 1.E+06
No. of cycles at peak phase angle
No.
of c
ycle
s at
tran
sitio
n po
int
AAD
AAM
AAD+LS
AAD+HL
AAM+HL
BASE
AirBlown
SBS
EVA
WIT
SAG
HCR
Line of Equality
Cross plot of No. of cycles to peak phaseangle and Nt
![Page 11: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/11.jpg)
( ) ( ) ( ) 1max1
2max
*
21*
)(
)(
B
D
kf
f AG
CCG
Ik
DfN −−−
=
⎟⎟
⎠
⎞
⎜⎜
⎝
⎛= γγω
ωπ
αα
α
( ) Bf AN −= maxγ
comparable to phenomenological regression model
Mechanistic Fatigue Model
![Page 12: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/12.jpg)
Comparing mechanistic parameters
{
{PM
HL
![Page 13: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/13.jpg)
1.E+02
1.E+03
1.E+04
1.E+05
1.E+06
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06
Measured fatigue life
Pred
icte
d fa
tigue
life
AAD
AAMHCR
AAD+LSAAD+HL
AAM+HLBASE
AirBlown
SBSEVA
WITSAG
Line of Equality
Cross plot between predicted and measuredfatigue life
![Page 14: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/14.jpg)
Test Method
• Bending beam test (asphalt mixes)
• Repeated load indirect tension test (asphalt mixes)
• DMA / DSR (asphalt mastics)
Test Mode
• Controlled stress
• Controlled strain
Analysis Method
• Direct methods
• Energy methods
• Crack growth modeling
• Rest periods (recovery)
![Page 15: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/15.jpg)
Analysis Methods : Energy Methods
• Direct methods based on number of cycles to a specific failure criteria is reached are empirical and are impacted by external factors such as applied stress level and sample geometry
• Energy methods should account for stress level and mode of loading provided the dissipated energy due to damage is correctly calculated
• Ideally, dissipated energy due to damage is computed after correcting for effects of viscoelasticity and non linearity
![Page 16: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/16.jpg)
Analysis Methods : Energy Methods
• Energy lost due to fracture in perfectly elastic material is easily quantified from hysteresis in stress-strain domain
• For viscoelastic materials hysteresis is also due to viscoelastic relaxation - not only damage
• Viscoelastic effect is eliminated by using stress-pseudo strain domain (based on Schapery’s method) in lieu of stress-strain domain
![Page 17: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/17.jpg)
Analysis Methods : Energy Methods
Illustration of undamaged viscoelastic material response in stress – strain domain vs. stress – pseudo strain domain
-60
-40
-20
0
20
40
60
0 5 10 15 20 25 30
Stress
Stra
in /
Pseu
do S
train
Strain DomainPseudo Strain Domain
![Page 18: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/18.jpg)
Analysis Methods : Energy Methods
• For a viscoelastic material where damage is occurring, stress –pseudo strain domain records the hysteresis used to monitor damage - referred to as DPSE
• Damage can be computed and summed until failure (CDPSE) as a measure of fatigue cracking resistance
![Page 19: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/19.jpg)
Analysis Methods : Energy Methods
-150000
-100000
-50000
0
50000
100000
150000
-0.005 -0.004 -0.003 -0.002 -0.001 0 0.001 0.002 0.003 0.004 0.005
Pseudostrain
Stre
ss
2000 Cycles30000 Cycles45000 Cycles Near Failure
Illustration of damaged viscoelastic material response in stress – pseudo strain domain (controlled-strain)
![Page 20: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/20.jpg)
Analysis Methods : Energy MethodsIllustration of use of DPSE and CDPSE to characterize fatigue cracking of mastics
0.0E+00
5.0E+09
1.0E+10
1.5E+10
2.0E+10
2.5E+10
1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05
No. of loading cycles
DPS
E
05000
1000015000200002500030000350004000045000
1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05
No. of loading cycles
Cum
ulat
ive
DPS
E (*1
0^10
)
![Page 21: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/21.jpg)
Test Method
• Bending beam test (asphalt mixes)
• Repeated load indirect tension test (asphalt mixes)
• DMA / DSR (asphalt mastics)
Test Mode
• Controlled stress
• Controlled strain
Analysis Method
• Direct methods
• Energy Methods
• Crack growth modeling
• Rest periods (recovery)
![Page 22: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/22.jpg)
Analysis Methods : Crack Growth Modeling
• Direct methods are empirical
• Energy methods are better than direct methods and can characterize fatigue cracking in mastic samples, BUT in order to predict crack growth in a mix a model that unifies various material properties is required
• Crack growth modeling is a powerful tool since it can bring together fundamental material properties such as surface energy and undamaged modulus into a single parameter
![Page 23: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/23.jpg)
DMA Mechanistic Approach
[ ]nRdr A JdN
=
n = 1+1/m Tensile strength is constant
n = 1/m Bond energy and α are constants
m = is the exponent in the relaxation modulus equation
mtEEtE −∞ += 1)(
α
α(N) increases for stress controlled and decreases or is constant for strain controlled
![Page 24: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/24.jpg)
PARAMETERS
Dynamic Mechanical Analyzer
Relaxation Test
Cyclic load at low strain
Cyclic load at high strain
Wilhelmy Plate
Universal Sorption Device
Surface energy components of asphalt binder
TEST DEVICE
Surface energy components of aggregate
bFrom relationship of DPSE vs. N
GR -Undamaged Reference Modulus
E1 and m
By fitting E(t) = E0 + E1tm
TEST TYPE
ΔGf Adhesive or Cohesive fracture energy
ΔR(N) – Crack Growth Index = fn(E1, m, GR, b, ΔGf)
Analysis Methods : Crack Growth Modeling
![Page 25: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/25.jpg)
Surface Energy Measurement
Asphalt: Low energy surfaceContact Angle Method
Aggregate: High energy surfaceVapor Adsorption Method
![Page 26: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/26.jpg)
Surface EnergyWilhelmy Plate Test Adsorption Test (USD)
Output: Contact angles Output: Adsorption isotherm
Analysis: Work of adhesion with probe liquids
Analysis: Work of adhesion with probe vapors
Result: Three surface energy components
Result: Three surface energy components
Performance related parameters:ΔGf; ΔGAS; ΔGWAS
Fracture potential and moisture sensitivity of mastics and mixtures
![Page 27: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/27.jpg)
Applications
![Page 28: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/28.jpg)
Application: Crack Growth Modeling
ΔR @ 50,000 cycles
ΔGf is adhesive bond energy
ΔGf is cohesive bond energy
Avg. St. Dev. Avg. St. Dev.
A Good 6.5 0.54 5.9 0.57B Intermediate 10.4 2.24 7.6 1.17C Fair 15.1 1.34 11.1 1.13
MixFatigue cracking reported
from qualitative field observations
Example of Application of the Model from a Recent Study
![Page 29: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/29.jpg)
Test Method
• Bending beam test (asphalt mixes)
• Repeated load indirect tension test (asphalt mixes)
• DMA / DSR (asphalt mastics)
Test Mode
• Controlled stress
• Controlled strain
Analysis Method
• Direct methods
• Energy Methods
• Crack growth modeling
![Page 30: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/30.jpg)
Application: Comparing Controlled Stress vs. Controlled Strain
• Classical views (Tangella et al. 1990):
Test ModeParameter Controlled
StressControlled
StrainAsphalt concrete thickness Thick ThinFailure criteria Complete failure Not well establishFatigue life Lower HigherSensitivity to mixture variables Higher LowerRate of dissipated energy Higher LowerHealing effect due to rest periods More beneficial Less beneficial
![Page 31: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/31.jpg)
Application: Controlled Stress vs. Controlled Strain
• Empirical criteria such as load cycles to failure, or load cycles to 50% reduction in modulus are inappropriate to unify the two approaches (Benedetto et al. 2004)
• Criteria based on energy methods and crack growth model show promising results in unifying the two approaches (Kim et al. 1997; Lee and Kim, 1998; Ghuzlan and Carpenter 2000; Masad et al. 2006)
• The total energy to failure should be consistent irrespective of the mode of loading provided failure is well defined in both modes and the dissipated energy due to damage is calculated correctly
![Page 32: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/32.jpg)
Application: Controlled Stress vs. Controlled Strain
• Crack growth modeling approach described earlier can be used to unify control stress and control strain
• The key is the function that relates ΔR(N) to E1, m, GR, b, ΔGfis different depending on the mode of loading
![Page 33: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/33.jpg)
Application: Controlled Stress vs. Controlled Strain
0
2
4
6
8
10
12
14
16
0 2 4 6 8 10 12 14 16
ΔR(N) Controlled Strain
ΔR(N
) Con
trolle
d S
tress
Case 1Equality Line
Example of unifying results from controlled stress and controlled strain using crack growth modeling
![Page 34: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/34.jpg)
Comparison of Results
Mode of Loading
Controlled Strain Controlled Stress
Average SD
Coef. of Variation
(%) Average SD
Coef. of Variation
(%)
Fatigue Life 69,000 51.41 74.51 139,600 62,870 45.04
Cumulative Dissipated Energy 7.60E+07 5.97E+07 78.66 3.74E+07 1.57E+07 42.04
ΔR(N), N=50,000 5.79 0.32 5.58 4.52 0.84 18.60
ΔR(N)/ln(N) 0.54 0.03 5.38 0.47 0.05 10.27
Parameter
Close for both modes of loading
![Page 35: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/35.jpg)
Application: Evaluating healing effect - HPI
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0 5000 10000 15000No. of loading cycles
Nor
mal
ized
DM
without RPwith RP
Slope = A
Slope = B
HPI = (A-B)/A
![Page 36: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/36.jpg)
Application: Evaluating effect of rest periods(Fatigue life increase)
0.E+00
1.E+10
2.E+10
3.E+10
4.E+10
5.E+10
0 2000 4000 6000 8000 10000 12000 14000
No. of loading cycles
DM
* N
o. o
f loa
ding
cyc
les
without RP
with RP
![Page 37: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/37.jpg)
Application: Comparing healing potential
01020304050607080
AAD-1 AAM-1 HCR HCR-A ABD AAB
HPIFLI
Average, %
35 ergs/cm2
49 ergs/cm2 65 ergs/cm2
27 ergs/cm2
![Page 38: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/38.jpg)
Application: Comparing healing potential
0
5
10
15
20
25
30
35
AAD-1 AAM-1 ABD AAB
% Change in ΔR Due to RP
35 ergs/cm2
49 ergs/cm2
65 ergs/cm2
27 ergs/cm2
![Page 39: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/39.jpg)
Application: Comparing influence of level of damage on percent recovery in DPSE due to healing
0
5
10
15
20
25
30
35
AAD-1 AAM-1 ABD AAB
LowHigh
% Recovery
35 ergs/cm2
49 ergs/cm265 ergs/cm2
27 ergs/cm2
![Page 40: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/40.jpg)
Application: Comparing increase in healing potential when number of rest periods is doubled
05
1015202530354045
AAD-1 AAM-1 ABD AAB
Binder
Average, % [(HPI2 – HPI1)/HPI1]
35 ergs/cm2
49 ergs/cm2
65 ergs/cm2
27 ergs/cm2
![Page 41: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/41.jpg)
Carpenter and Shen Study (TRB 2006)
• Used ratio of dissipated energy approach• With inclusion of RP of 9 seconds, fatigue
life increased by factor of 10• Healing at low damage condition is key to
understanding endurance limit• Healing in polymer-modified asphalts is
superior
![Page 42: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/42.jpg)
Application: Moisture damageWilhelmy Plate Test Adsorption Test (USD)
Output: Contact angles Output: Adsorption isotherm
Analysis: Work of adhesion with probe liquids
Analysis: Work of adhesion with probe vapors
Result: Three surface energy components
Result: Three surface energy components
Performance related parameters:(1) ΔGAS (2) ΔGWAS
Moisture Sensitivity of Mixes:Field / Laboratory Performance
![Page 43: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/43.jpg)
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
log( ER*SSA)
log
(Mod
ulus
Rat
io)
RARKRLLinear FitC.I. for Fit 95%
Tensile Modulus Ratio (wet to dry) v. ER*SSA
R2 = 0.85
![Page 44: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/44.jpg)
Asphalt and aggregates can be individually tested to generate an array of values for all
possible mixture combinations
Granite A
Gravel A
Gravel B
Limestone A
Limestone B
Quartzite A
Sandstone A
Sandstone B KAG NAG
PG 64-22 A 0.40 1.05 0.58 0.53 0.74 0.56 1.37 0.88 0.35 0.63
PG 64-22 B 0.65 1.98 1.00 0.86 1.38 0.97 2.54 1.41 0.71 1.08
PG 64-22 C 0.49 1.48 0.76 0.66 1.04 0.74 1.90 1.10 0.51 0.82
PG 64-22 D 0.58 1.38 0.80 0.72 1.03 0.78 1.69 1.10 0.58 0.86
PG 64-22 E 0.61 1.96 0.97 0.83 1.34 0.94 2.57 1.40 0.66 1.05
PG 64-22 F 0.74 2.47 1.18 1.00 1.66 1.14 3.29 1.71 0.81 1.27
PG 64-22 G 0.62 1.92 0.97 0.83 1.34 0.94 2.47 1.37 0.68 1.04
PG 64-22 H 0.47 1.26 0.69 0.62 0.90 0.67 1.63 1.02 0.45 0.75
PG 64-22 I 0.69 2.20 1.08 0.92 1.51 1.05 2.87 1.55 0.75 1.17
PG 64-28 B 0.46 0.93 0.59 0.56 0.70 0.58 1.16 0.83 0.40 0.64
PG 76-22 A 0.72 1.49 0.93 0.87 1.15 0.92 1.79 1.24 0.70 1.00
PG 76-22 B 0.45 1.07 0.62 0.57 0.78 0.61 1.37 0.91 0.40 0.68
PG 76-22 C 0.67 1.24 0.83 0.79 0.99 0.82 1.48 1.09 0.62 0.89
PG 76-22 D 0.60 1.50 0.85 0.77 1.10 0.83 1.88 1.19 0.60 0.92
![Page 45: NANO ÆMICRO ÆMESO ÆMACRO · 2009. 3. 4. · 2.0E+07 4.0E+07 6.0E+07 8.0E+07 1.0E+08 0 2000 4000 6000 8000 10000 No. of loading cycles Nonlinear dynamic modulus (Pa) 20 25 30 35](https://reader036.fdocuments.net/reader036/viewer/2022071405/60face7bb5dd076c9863a715/html5/thumbnails/45.jpg)
Fast, repeatable, many replicate samplesAnalysis: direct, DE, fracture mechanics (including bond energy)Monitor of effects of additives, fillersMonitor of healing potentialBasis for unifying controlled-stress and controlled-strain