Regional Implementation of the MSCR Test Pennsylvania Asphalt Pavement Association 54th Annual Conference Hershey, PA January 30, 2014 Gregory A. Harder, P.E.

Past MSCR Presentations Development of Standard Practice for Superpave Plus Specifications - J. D'Angelo – 2005 – Burlington, VT Binder and Mixture ETG Update - J. D'Angelo – 2006 – Wilmington, DE FHWA Pavements Program - What's Happening - J. D'Angelo – 2007 – Mystic, CT Current Status for Multiple Stress Creep Recovery October 9, 2008 - K. Mooney – 2008 – Atlantic City, NJ Asphalt Mixture and Binder Expert Task Group Update - J. Bukowski – 2008 – Atlantic City, NJ Update on National Issues - ETG Activity - J. Bukowski – 2009 – Portland, ME MSCR Test - A New High Temp Spec - J. D'Angelo – 2009 – Portland, ME Update on National Issues - ETG Activity - J. Bukowski – 2010 – Saratoga, NY Binder Grade Selection Using the MSCR Specification - J. D'Angelo – 2010 – Saratoga, NY Update on National Issues - ETG Activity - J. Bukowski – 2011 – Providence, RI Regional Implementation of the MSCR Test – G. Harder – 2012 – Philadelphia, PA

MSCR Implementation The use of polymer modified binders has grown tremendously over the past several years However, the most widely used binder specification in the U.S., AASHTO M 320, was based on a study of neat (unmodified) binders, and may not properly characterize polymer modified binders

Why doesn’t M 320 properly characterize polymer-modified binders? Current spec, G* and δ are measured in the linear visco-elastic range. For neat binders, flow is linear(strain increases in a constant proportion to stress) and therefore not sensitive to the stress level of the test. For polymer-modified binders, the response is not linear and very sensitive to the stress level of the test. The polymer chains can be rearranged substantially as the stress increases. The slide show continues with the second and third bullet items. Neat asphalts behave in a linear fashion, where the strain increases in a constant proportion to the stress. In polymer-modified binders, the strain does not always increase in a linear fashion. At higher stress levels, the polymer chains can be substantially rearranged, resulting in non-linear behavior.

PG Grading Alone Does Not Always Predict Performance Study of the two mixes with the same aggregate structure, but different binders. PG 63-22 modified, no rutting PG 67-22 unmodified, 15mm rut Furthermore, PG grading alone does not predict performance, especially when comparing neat and modified binders.

Why Do We Need New Binder Test? PG Binders Most Common “Neat” Binder Grades PG 64-22 PG 67-22 Most Common “Modified” Binder Grade PG 76-22 Works OK for neat binders Doesn’t work as well for modified binders Like I said previously, the SEAUPG states decided to adopt the Performance Grade (PG) binder grading system back in the mid 1990s. Each of the states decided to adopt a series of binder grades for use in their states, depending on various factors. Probably the most common grades used in the SE are as follows: the most common “neat” or unmodified binders were PG 64-22 and PG 67-22. the most common “modified” grade is PG 76-22 Over the last 17 years, the PG system has served us pretty well. There is no doubt that we have come a long ways with this approach. So how does the specification working…click…the current PG grading system works well for the neat binders. However…click…experience and research has shown that the current PG does not work as well for modified binders. This was recognized from the start so many states decided to adopt additional tests (commonly referred to as “PG Plus” tests) like Elastic Recovery, phase angle or force ductility to ensure that the modifiers were present in the binder.

What happened as a result of M 320’s inability to fully characterize polymer-modified binders? Most states began requiring additional tests to the ones required in AASHTO M 320 These mostly empirical tests are commonly referred to as “PG Plus” tests These tests are not standard across the states – difficult for suppliers Even some of the tests that are the most common, e.g. Elastic Recovery, are not run the same way from state to state

States with a “PG Plus” Specification No PG Plus Spec

ER Information and Test Time The Elastic Recovery Test is an excellent tool to establish the presence of polymer modification. It takes about 4 hours to prepare and test samples for this information. However, it is a poor tool to evaluate the rutting performance of polymer-modified binders. The MSCR test can use the same sample already being run in the DSR to give more information in a few extra minutes. ER - Pour, cool, trim, load, pull, wait, measure

Multiple Stress Creep Recovery Test Performed on RTFO-aged Binder Test Temperature Environmental Temperature Not Grade-Bumped 10 cycles per stress level 1-second loading at specified shear stress 0.1 kPa 3.2 kPa 9-second rest period

Multiple Stress Creep Recovery Test Calculate Recovery for each Cycle, Stress Difference between strain at end of recovery period and peak strain after creep loading Calculate Non-recoverable Creep Compliance (Jnr) Non-recoverable shear strain divided by applied shear stress “J” = “compliance” “nr” = “non-recoverable”

ALF Study - 7 Asphalt Binders AZ CRM ---- 70-22 PG 70-22 Control Air Blown SBS TX TBCR TP PG 70-22 + Fibers PG 70-22 SBS 64-40 Air Blown SBS TP An Accelerated Load Facility (ALF) at Turner-Fairbanks Highway Research Center conducted rut testing on mixes incorporating 7 different binder types. They then tried to correlate the binder performance as measured by Superpave binder testing and MSCR testing to the observed rutting. 1 2 3 4 5 6 7 8 9 10 11 12

ALF Loading The pavement was heated to a constant 64ºC. The FHWA ALF uses an 18,000 lbs wheel load with no wheel wander. The speed is 12 MPH. This is a extreme loading condition far more sever than any actual highway.

Relationship between G*/ sinδ and ALF rutting As you know, the coefficient of determination, R2, is a measure of how well future outcomes can be predicted by a model, “0” being unable to reliably predict performance and “1” being a perfect predictor of performance. The Superpave binder grading, at an R2 of 0.1261, is a poor predictor of modified binder rutting in this ALF study. Existing SHRP specification has poor relationship to rutting for modified systems.

Relationship between Jnr and ALF rutting 25.6kPa MSCR can adjust for field conditions and has excellent relations to performance. On the other hand, the MSCR Jnr, at an R2 of 0.8167, is a reliable predictor of modified binder rutting in this ALF study.

Miss I-55 6yr rut Jnr 3.2 kPa

New PG Grading System (MSCR) AASHTO MP-19 Environmental grade plus traffic level designation; i.e. PG 64E-22 Based on Climatic Temperature High and Low Pavement Temperature Traffic Designation “S” – Standard: < 10 million ESALs and standard traffic loading “H” – Heavy: 10 – 30 million ESALs or slow moving traffic “V” – Very Heavy: > 30 million ESALs or standing traffic “E” – Extreme: > 30 million ESALs and NO MORE GRADE BUMPING

New PG Grading System (MSCR) Requirements S = Standard: Jnr ≤ 4.5 kPa-1 H = Heavy: Jnr ≤ 2.0 kPa-1 V = Very Heavy: Jnr ≤ 1.0 kPa-1 E = Extr. Heavy: Jnr ≤ 0.5 kPa-1

AASHTO M320 Grades As an example, following Table 1… Kentucky’s climate requires a PG 64-22 to meet high and low pavement design temperatures PG 64-22 is a standard, unmodified asphalt binder grade For Very Heavy (slow and/or very high volume) traffic, Kentucky “grade bumps” to a PG 76-22 asphalt binder PG 76-22 is a modified asphalt binder grade Kentucky requires ER ≥ 75%

AASHTO M320 Grades As an example, following Table 1… Although a PG 76-22 is specified, Kentucky recognizes that the pavement temperature never gets that hot Original DSR (T315): G*/sin d ≥ 1.00 kPa @ 76°C RTFO DSR (T315): G*/sin d ≥ 2.20 kPa @ 76°C Grade bumping is a way to get a stiffer asphalt binder at the expected high pavement temperature

AASHTO MP-19 Grades As an example, following MP-19… Kentucky’s climate requires a PG 64-22 to meet high and low pavement design temperatures PG 64-22S is an unmodified asphalt binder grade that can be used for Standard (normal speed, normal volume) traffic conditions For Very Heavy (slow and/or very high volume) traffic, Kentucky requires a PG 64-22V asphalt binder PG 64-22V is a modified asphalt binder grade Particularly if MSCR Recovery requirement is added

AASHTO MP-19 Grades As an example, following MP-19… Instead of increasing the test temperature as in M-320, Kentucky recognizes that the pavement temperature never gets that hot and adjusts criteria Original DSR (T315): G*/sin d ≥ 1.00 kPa @ 64°C RTFO MSCR (TP70): Jnr ≤ 1.0 kPa-1 @ 64°C Get a stiffer asphalt binder at the expected high pavement temperature

NEAUPG Tasked by steering committee Formed Implementation Team Mission statement, description, target audience, opportunities, obstacles, strategies, and goals identified Performed round robin testing All states participated in pooled fund, each will be obtaining/using the same Malvern DSR States agreed to specify both MP 19 Jnr requirements and MP 70 MSCR Recovery requirements beginning January 2014 (H, V, and E grades only – S grades remain M320) NY MP 19 on all grades including S, recovery H, V, and E grades only

MSCR Implementation Team

Goals

MSCR % Recovery validates polymer modification AASHTO TP 70 % Recovery above the line means asphalt binder is modified with an acceptable elastomeric polymer Plotting the % recovery can show the presence of an elastomeric polymer, as outlined in Appendix X2 of AASHTO TP 70. Below line - not modified with an elastomeric polymer 26

MODIFIED GRADES IN THE NORTHEAST CURRENT GRADE – M320 NEW GRADE – MP-19 PG 76-22 PG 64E-22 PG 76-28 PG 64E-28 PG 70-28 PG 64V-28 PG 64-22P PG 64V-22 PG 58-34 PG 58H-34

WHAT IS CURRENTLY BEING SUPPLIED?

MEDOT BASE – 58C CURRENT GRADE – M320 NEW GRADE – MP-19 PG 58-28 PG 58S-28 PG 58-34 PG 58H-34 PG 64-28(PPA) PG 58H-28 PG 70-28 PG 58V-28 PG 76-28 PG 58E-28 Will be specifying for modified grades on Jan. 1, 2014 – PG 58V-28[PG 70-28] – little to no use of PG 58-34 – not sure yet on PG 64-28 (PPA modified)

NHDOT BASE – 58C CURRENT GRADE – M320 NEW GRADE – MP-19 PG 64-28 PG 58H-28 PG 64-28(PPA) PG 70-28 PG 58V-28 PG 76-28 PG 58E-28 Use a very limited amount of modified grades but will implement PG 58V and PG 58E grades on Jan. 1, 2014

NJDOT BASE – 64C CURRENT GRADE – M320 NEW GRADE – MP-19 PG 64-22 PG 64S-22 PG 76-22 PG 64E-28 Will be specifying MSCR for PG 76-22 effective Jan.1, 2014 – not sure if it will be PG 64V-22 or PG 64E-22

Will implement on Jan. 1, 2014 if suppliers want to DelDOT BASE – 64C CURRENT GRADE – M320 NEW GRADE – MP-19 PG 58-28 PG 58S-28 PG 64-22 PG 64S-22 PG 70-22 PG 64H-22 PG 76-22 PG 64E-22 Will implement on Jan. 1, 2014 if suppliers want to

MDSHA BASE – 64C CURRENT GRADE – M320 NEW GRADE – MP-19 PG 58-22 PG 58S-22 PG 58-28 PG 58S-28 PG 64-22 PG 64S-22 PG 64-28 PG 64H-28 PG 70-22 PG 64H-22 PG 76-22 PG 64E-22 Working on software upgrade to test for MSCR – will collect data for information – implementation of MSCR possible in 2014

VT-AOT BASE – 58C CURRENT GRADE – M320 NEW GRADE – MP-19 PG 52-34 PG 52S-34 PG 58-28 PG 58S-28 Currently running MSCR on 75% of the full set testing – base temperature will be 58C – but will test at 52C for those mixtures containing high RAP – all projects for 2014 have been bid – have been using MSCR on emulsion residue

CTDOT BASE – 58C CURRENT GRADE – M320 NEW GRADE – MP-19 PG 58-28 PG 58S-28 PG 64-22 PG 64S-22 PG 76-22 PG 64E-22 Limited amount of PG 76-22 polymer modified is used - currently running test and collecting data – considering allowing substitution of MP-19 grades in 2014 with full implementation for all grades possible in 2015

Mass Highway BASE – 58C CURRENT GRADE – M320 NEW GRADE – MP-19 PG 52-34 PG 52S-34 PG 64-28 PG 64S-22 or PG 58H-28 PG 64-28 Rubber PG 64?-28 PG 64-28 SBR Currently discussing internally – want to talk with industry this winter as to how to move forward – no implementation on Jan.1 – hope to have better idea of direction in early 2014

NYSDOT BASE – 58/64C CURRENT GRADE – M320 NEW GRADE – MP-19 PG 58-34 PG 58H-34 PG 64-22 PG 64S-22 PG 64-22P PG 64V-22 PG 70-22 PG 64H-22 PG 76-22 PG 64E-22 Full implementation MP-19 for all grades in 2014 – actual date yet to be determined

Already implemented PG 64V-28 and PG 64E-28 RIDOT BASE – 64C CURRENT GRADE – M320 NEW GRADE – MP-19 PG 64-28 PG 64S-28 PG 70-28 PG 64V-28 PG 76-28 PG 64E-28 Already implemented PG 64V-28 and PG 64E-28

PennDOT BASE – 58/64C CURRENT GRADE – M320 NEW GRADE – MP-19 PG 58-28 PG 58S-28 PG 64-22 PG 64S-22 PG 76-22 PG 64E-22 Have been collecting data - will allow a substitution of PG 64E-22 for PG 76-22 in 2014

Binder Grade Substitution PG 64E-22 = PG 76-22

Thanks