Low Temperature Cracking & HPAC Bill Buttlar, Glen Barton Chair University of Missouri-Columbia

HPAC Workshop ETS, Montreal 9/30/2016

Outline

• Introduction • DC(T) Low-Temp Fracture Test Dev. • ILLI-TC Model Dev. • Hamburg-DC(T) Plot Method • HPAC Low-Temp Project Examples • Recommendations for HPAC for Low Temps

3

Mizzou Asphalt Pavement Innovation Lab (MAPIL)

MAPIL

Educational Programs

Industry Partners

Agency Partners

Academic Partners

Other Partners

Affiliate Labs

4

Mizzou Asphalt Pavement Innovation Lab (MAPIL)

MAPIL

Superpave Binder Lab Superpave Mix Design Lab

Sustainable Asphalt Lab Infrastructure Advanced/Innovative Performance Testing Lab – DC(T), IDT, Hamburg, Universal Test Machine, Advanced

Binder Tests (DMA, FTIR), Innovation/Maker Lab Area

Agency Partners MoDOT, Cities, FHWA, NCHRP, OMP/Airports, FAA, LRS, NSF,

AF, Army, IFSTTAR, EMPA

Industry Partners MAPA, NAPA, Asphalt

Institute, Road Science, MWV, Marathon, Seneca, Kraton, HRG, Southwind

RAS, Reliable, Saint-Gobain, Agg. Prod.,

Colas, Troxler, Test Quip, HMA Lab Supply, Instrotek

Academic Partners Rolla, UIUC, TAMU, GT, UT, Arkansas, Iowa State, UM, MTU, PSU, UC Davis, UMass,

Auburn, KTH, URI, Nottingham, ENTPE, TU Delft, Parma

Educational Programs Senior/grad Asphalt Materials and Design, Grad Advanced Asphalt/Research, Online

Course/Training/Professional Education, Collaboration with Other Courses, Educational Thrusts (Sustainable Infrastructure Materials, Smart Cities, New Materials), Maker Lab

Affiliate Labs Concrete, Soils,

Geosynthetics, Materials Science, NDT,

Composite Materials, Chemistry, Trans.

Systems, Smart City, Computational

Mechanics, Simulation

Other Partners City Digital, National Labs, Ill. State Toll Hwy Auth., NCAT,

WRI

Rutting – Under Control w/ Hamburg Rut Test ……

0

2

4

6

8

10

12

14

16

180 5000 10000 15000 20000

Rut D

epth

(mm

)

Number of Wheel Passes

Example Hamburg Profiles

Passing

Failing

12.5mm Max. Rut Depth

500C

6

….Cracking….not so much…..

- If its not durable, it’s probably not sustainable

Reflective

Block Reflective/Thermal

Disk-Shaped Compact Tension - DC(T)

Fracture Plane Motivation – measure fracture

energy, use cylindrical specimens, maximize

repeatability, use true fracture test

Based on ASTM E399 – Geometry slightly modified to account for differences in the fracture behavior of steel and

asphalt concrete

Genesis was NSF GOALI study on reflective cracking: UIUC-

NSF-Koch (2004) Wagoner, M. P., Buttlar, W. G., and G. H. Paulino, “Disk-Shaped Compact Tension Fracture Test: A Practical Specimen Geometry for Obtaining Asphalt Concrete Fracture Properties,” Experimental Mechanics, Vol. 45, No. 3, pp. 270-277, 2005.

Induced Displacement via Steel Loading Pins

8

CMOD Clip Gage Spring Mounted onto

Knife-Edge Gage Points

CMOD = Crack Mouth Opening Displacement

Early DC(T) Test at U.

of Illinois

9

ASTM Specification circa 2006

DC(T) Testing

• Test time: less than 10 minutes • Load specimen, then turn-key

operation • ~ $49k device • 110V wall outlet Test Quip DC(T)

DC(T) Results from Pooled Fund Study

11

PGLT + 10oC

Field-Aged Cores (Assumed Long-Term Aged)

SCB also evaluated, but found by Univ. of MN to have high COV and poor correlation to field cracking in blind study

12

New DC(T) Based Thermal Cracking Spec

From: http://www.cts.umn.edu/Publications/ResearchReports/reportdetail.html?id=2178 Implementation: Minnesota, Iowa, Wisconsin, Chicago DOT, O’Hare

Low Temperature Cracking

TC Model Stress Intensity Factor

Paris ‘Law’

Crack amount model

Old TC Model vs. ILLI-TC

13

ILLI-TC Finite element based thermal

cracking prediction model with cohesive zone modeling

Stress Intensity Factor Far-field stress at depth of crack

Current crack length

)99.145.0( 56.00CK +=σ

Change in crack depth

Change in stress intensity factor Fracture parameters

nKAC )(∆=∆

Amount of cracking is a function of the probability that the crack depth is equal to or greater the thickness of the surface layer

E1 E5

τ1 τ5

Low Temperature Cracking

Bilinear Cohesive Zone Model

14

Cracking Cohesive Zone (Softening/Damage)

True crack tip Cohesive crack tip

δc

σt

tσ

Cδ

Fracture Energy = f(area)

Trac

tion

(MPa

)

Displacement Jump (mm)

Unloading

Reloading

Bilinear CZM (Song et al., 2006)

Low Temperature Cracking

FE Viscoelastic Formulation Recursive-incremental time integration scheme

15

( ) ( ) ( )'

'

'' '

'

,, ,

t t

t

x tx t C x dt

tε

σ ξ ξ=

=−∞

∂= −

∂∫ ( ) ( ) ( ) ( ), Rd d dξ ξ ξ ξ= × +σ K x ε σ

T(t)

x

y

Low Temperature Cracking 10/16/2016 16

Low Temperature Cracking 10/16/2016 17

ILLI-TC Pre-Analyzer Results

Low Temperature Cracking 10/16/2016 18

Thermal Stress in Longitudinal Direction near the Crack Path (MPa)

Thermal Stress in Longitudinal Direction near the Crack Path (MPa)

Thermal Stress in Longitudinal Direction near the Crack Path (MPa)

Example ILLI-TC Crack Simulation Results

Stability with Crack-Resistance: Two-Dimensional View of Performance

+

Hamburg DC(T)

“Performance-Space” Diagram 0

5

10

15

20

250 200 400 600 800 1000 1200

Ham

burg

Rut

Dep

th (m

m)

DC(T) Fracture Energy (J/m2)

Super Mix (SMA, heavy traffic)

Soft Mix (reflective crack control)

Poor Mix (Non-surface, low-traffic, or temporary use only)

Stiff Mix (Bottom layer of full-depth pavement)

Performance-Space Diagram: Zones

0

5

10

15

20

250 200 400 600 800 1000 1200

Ham

burg

Rut

Dep

th (m

m)

DC(T) Fracture Energy (J/m2)

Poor Mix, Failing Soft Mix, Fails Spec.

Passing, Low, Med. & High Traffic Levels

Stiff Mix, Failing

Passing, Low Traffic Only

Passing, Low & Medium Traffic

12.5

400 460 690

Softer Binder, No Polymer

0

5

10

15

20

250 200 400 600 800 1000 1200

Ham

burg

Rut

Dep

th (m

m)

DC(T) Fracture Energy (J/m2)

Softer binder did not help, Culprit: weak aggregate

Stiff Mix, Failing

PG 58-28

PG 64-22

Mix Adjustment: Binder Modification Hypothesized

0

5

10

15

20

250 200 400 600 800 1000 1200

Ham

burg

Rut

Dep

th (m

m)

DC(T) Fracture Energy (J/m2)

Highly Modified Grade

Modified, Harder Grade

Highly Modified Grade

Modified, Softer Grade

Mix Adjustment: Binder Modification

Mix Affects: RAS/Recycling

Mix Affects: RAP/Recycling

Mix Effects: Stronger Aggregate

Early Performance-Space Data for Illinois: How are we Doing?

0

5

10

15

20

250 200 400 600 800 1000 1200

Ham

burg

Rut

Dep

th (m

m)

DC(T) Fracture Energy (J/m2)

Not surprisingly,

tendency towards stiff

mixes

Stiff Mix, Failing

Passing

Low Med High Traffic

Poor Mix, Failing Soft Mix, Fails Spec.

SMA’s

0

5

10

15

20

250 200 400 600 800 1000 1200

Ham

burg

Rut

Dep

th (m

m)

DC(T) Fracture Energy (J/m2)

These three mixes are

SMA’s

Stiff Mix, Failing

Passing

Low Med High Traffic

Poor Mix, Failing Soft Mix, Fails Spec.

High ABR Mixes

0

5

10

15

20

250 200 400 600 800 1000 1200

Ham

burg

Rut

Dep

th (m

m)

DC(T) Fracture Energy (J/m2)

N90, 28.6% ABR, 2.4% RAS, High Polymer, WMA

N70, 29.8% ABR, 6.2% RAS, Soft Binder, WMA

N90, 28.6% ABR, 3.0% RAS, High Polymer

N30, 67% ABR, Very Soft Binder, WMA

7 Tollway SMA Surface Course Mixes – Placed 2008 – 2012, Sampled 2015

31

A

B C

D

E

F

G

Detail of SMA Surface Mixes Cored & Evaluated

32

Mix Location Year Placed AC Grade ABR % Surface

Thickness Coarse Agg. Type

A. I-90 WB near Rockford 2009 PG 76-22 GTR 14 2” Cr. Gravel

B. I-90 EB near Rockford 2008 PG 76-22 GTR 16 2” Diabase

C. I-90 EB near Newberg Rd 2009 PG 76-22 SBS 36* 2” Quartzite D. I-90 WB near Rt. 25 /

Elgin 2011 PG 70-28 SBS 33* 1.75” Quartzite

E. I-88 EB East of DeKalb 2012 PG 70-28 SBS 37* 1.5” Cr. Gravel

F. I-355 NB at 63rd St. 2009 PG 76-22 GTR 0 1.75” Steel Slag

G. I-294 NB, N. of Cermak 2012 PG 70-28 SBS 31* 2” Quartzite

* With RAS

ILLI-TC Thermal Cracking Modeling

33

Table 2. Critical events as predicted by ILLI-TC

Section Cores Location

Input Output

Fracture Energy (J/m2)

Peak Load (kN)

Calculated Tensile Strength (MPa)

Peak Tensile Stress (MPa)

Peak Tensile Stress/ Tensile strength

(%)

Critical Events

A I-90 WB in Rockford 1275 3.38 4.92 1.15 23.4 0

B I-90 EB in Rockford 1176 2.76 4.01 0.96 23.9 0

C I-90 EB near Newberg Rd 1003 3.61 5.25 3.53 67.2 0

D I-90 WB in Rt. 25 in Elgin 1340 4.10 5.96 1.09 18.3 0

E I-88 EB, East of DeKalb 1038 2.47 3.59 2.72 75.8 0

F I-355 NB at 63rd St. 1135 3.64 5.29 2.87 54.3 0

G I-294 NB, N. of Cermak Toll 1222 2.84 4.13 2.32 56.2 0

IL Tollway SMA: ILLI-TC Modeling

34

Section Cores Location

Input Output

Fracture Energy (J/m2)

Peak Load (kN)

Calculated Tensile Strength (MPa)

Peak Tensile Stress (MPa)

Peak Tensile Stress/ Tensile strength

(%)

Critical Events

A I-90 WB in Rockford 1275 3.38 4.92 1.15 23.4 0

B I-90 EB in Rockford 1176 2.76 4.01 0.96 23.9 0

C I-90 EB near Newberg Rd 1003 3.61 5.25 3.53 67.2 0

D I-90 WB in Rt. 25 in Elgin 1340 4.10 5.96 1.09 18.3 0

E I-88 EB, East of DeKalb 1038 2.47 3.59 2.72 75.8 0

F I-355 NB at 63rd St. 1135 3.64 5.29 2.87 54.3 0

G I-294 NB, N. of Cermak Toll 1222 2.84 4.13 2.32 56.2 0

Hamburg-DC(T) Plot: Tollway SMA 0

5

10

15

20

250 200 400 600 800 1000 1200

Ham

burg

Rut

Dep

th (m

m)

DC(T) Fracture Energy (J/m2)

I-90 WB in Rockford

I-90 EB in Rockford

I-90 EB near Newberg Rd

I-90 WB in Rt. 25 in Elgin

I-88 EB, East of DeKalb

I-355 NB at 63rd St.

I-294 NB, N. of Cermak Toll

Suggests use of softer binder, w/ same UTR (PG spread)

36

Ultra-high Fracture Energy Mixes for Reflective Crack Control: ORD 9R Project

Accelerated Pavement Study (ATLAS)

ORD Solution: Ultra-high fracture energy mixtures, 850 - 1,300 J/m2

Some Old, but Useful Thoughts on Thermal Stress from

Strategic Highway Research Program, circa 1993

Typical Pavement Temperature versus Time for PTI Section 38

Typical Pavement Temperature versus Depth for PTI Section 38

Typical Pavement Strain versus Time for PTI Section 38

Typical Pavement Strain versus Depth for PTI Section 38

Typical Pavement Stress versus Time for PTI Section 38 Typical Pavement Stress versus

Time for PTI Section 38

Typical Pavement Stress versus Depth for PTI Section 38

Finding: When >= 50mm below surface, thermal stresses < 50% Thus, can safely use higher modulus materials in cold climates

Recommendations

• For lower pavement layers in cold climates, stiffer mixtures can be employed: – Need to consider reflective cracking – Must still design against fatigue

• For surface mixes, use balanced design: – For high profile/innovative applications, use

test plus model (i.e., DC(T) plus ILLI-TC) – WMA+RAP+RAS+GTR/SMA Mixes can

perform well on surfaces in cold climates (RAS & GTR require experience)

45

Thank you for your attention!

Questions/Comments