NCHRP RPT 598

NATIONAL
NCHRP REPORT 598

COOPERATIVE
HIGHWAY
RESEARCH
PROGRAM
Performance-Related Tests
of Recycled Aggregates for Use
in Unbound Pavement Layers
TRANSPORTATION RESEARCH BOARD 2008 EXECUTIVE COMMITTEE*
OFFICERS
CHAIR: Debra L. Miller, Secretary, Kansas DOT, Topeka
VICE CHAIR: Adib K. Kanafani, Cahill Professor of Civil Engineering, University of California, Berkeley
EXECUTIVE DIRECTOR: Robert E. Skinner, Jr., Transportation Research Board
MEMBERS
J. Barry Barker, Executive Director, Transit Authority of River City, Louisville, KY
Allen D. Biehler, Secretary, Pennsylvania DOT, Harrisburg
John D. Bowe, President, Americas Region, APL Limited, Oakland, CA
Larry L. Brown, Sr., Executive Director, Mississippi DOT, Jackson
Deborah H. Butler, Executive Vice President, Planning, and CIO, Norfolk Southern Corporation, Norfolk, VA
William A.V. Clark, Professor, Department of Geography, University of California, Los Angeles
David S. Ekern, Commissioner, Virginia DOT, Richmond
Nicholas J. Garber, Henry L. Kinnier Professor, Department of Civil Engineering, University of Virginia, Charlottesville
Jeffrey W. Hamiel, Executive Director, Metropolitan Airports Commission, Minneapolis, MN
Edward A. (Ned) Helme, President, Center for Clean Air Policy, Washington, DC
Will Kempton, Director, California DOT, Sacramento
Susan Martinovich, Director, Nevada DOT, Carson City
Michael D. Meyer, Professor, School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta
Michael R. Morris, Director of Transportation, North Central Texas Council of Governments, Arlington
Neil J. Pedersen, Administrator, Maryland State Highway Administration, Baltimore
Pete K. Rahn, Director, Missouri DOT, Jefferson City
Sandra Rosenbloom, Professor of Planning, University of Arizona, Tucson
Tracy L. Rosser, Vice President, Corporate Traffic, Wal-Mart Stores, Inc., Bentonville, AR
Rosa Clausell Rountree, Executive Director, Georgia State Road and Tollway Authority, Atlanta
Henry G. (Gerry) Schwartz, Jr., Chairman (retired), Jacobs/Sverdrup Civil, Inc., St. Louis, MO
C. Michael Walton, Ernest H. Cockrell Centennial Chair in Engineering, University of Texas, Austin
Linda S. Watson, CEO, LYNX–Central Florida Regional Transportation Authority, Orlando
Steve Williams, Chairman and CEO, Maverick Transportation, Inc., Little Rock, AR
EX OFFICIO MEMBERS
Thad Allen (Adm., U.S. Coast Guard), Commandant, U.S. Coast Guard, Washington, DC
Joseph H. Boardman, Federal Railroad Administrator, U.S.DOT
Rebecca M. Brewster, President and COO, American Transportation Research Institute, Smyrna, GA
Paul R. Brubaker, Research and Innovative Technology Administrator, U.S.DOT
George Bugliarello, Chancellor, Polytechnic University of New York, Brooklyn, and Foreign Secretary, National Academy of Engineering,
Washington, DC
J. Richard Capka, Federal Highway Administrator, U.S.DOT
Sean T. Connaughton, Maritime Administrator, U.S.DOT
LeRoy Gishi, Chief, Division of Transportation, Bureau of Indian Affairs, U.S. Department of the Interior, Washington, DC
Edward R. Hamberger, President and CEO, Association of American Railroads, Washington, DC
John H. Hill, Federal Motor Carrier Safety Administrator, U.S.DOT
John C. Horsley, Executive Director, American Association of State Highway and Transportation Officials, Washington, DC
Carl T. Johnson, Pipeline and Hazardous Materials Safety Administrator, U.S.DOT
J. Edward Johnson, Director, Applied Science Directorate, National Aeronautics and Space Administration, John C. Stennis Space Center, MS
William W. Millar, President, American Public Transportation Association, Washington, DC
Nicole R. Nason, National Highway Traffic Safety Administrator, U.S.DOT
Jeffrey N. Shane, Under Secretary for Policy, U.S.DOT
James S. Simpson, Federal Transit Administrator, U.S.DOT
Robert A. Sturgell, Acting Administrator, Federal Aviation Administration, U.S.DOT
Robert L. Van Antwerp (Lt. Gen., U.S. Army), Chief of Engineers and Commanding General, U.S. Army Corps of Engineers, Washington, DC
*Membership as of January 2008.

NATIONAL COOPERATIVE HIGHWAY RESEARCH PROGRAM
NCHRP REPORT 598

Athar Saeed
APPLIED RESEARCH ASSOCIATES, INC.
Vicksburg, MS
Subject Areas
Pavement Design, Management, and Performance • Materials and Construction
Research sponsored by the American Association of State Highway and Transportation Officials
in cooperation with the Federal Highway Administration
TRANSPORTATION RESEARCH BOARD

WASHINGTON, D.C.
2008
www.TRB.org
NATIONAL COOPERATIVE HIGHWAY NCHRP REPORT 598
RESEARCH PROGRAM
Systematic, well-designed research provides the most effective Project 4-31
approach to the solution of many problems facing highway ISSN 0077-5614
administrators and engineers. Often, highway problems are of local ISBN: 978-0-309-099202
interest and can best be studied by highway departments individually Library of Congress Control Number 2008900695
or in cooperation with their state universities and others. However, the © 2008 Transportation Research Board
accelerating growth of highway transportation develops increasingly
complex problems of wide interest to highway authorities. These
problems are best studied through a coordinated program of COPYRIGHT PERMISSION
cooperative research.
Authors herein are responsible for the authenticity of their materials and for obtaining
In recognition of these needs, the highway administrators of the written permissions from publishers or persons who own the copyright to any previously
American Association of State Highway and Transportation Officials published or copyrighted material used herein.
initiated in 1962 an objective national highway research program Cooperative Research Programs (CRP) grants permission to reproduce material in this
employing modern scientific techniques. This program is supported on publication for classroom and not-for-profit purposes. Permission is given with the
understanding that none of the material will be used to imply TRB, AASHTO, FAA, FHWA,
a continuing basis by funds from participating member states of the
FMCSA, FTA, or Transit Development Corporation endorsement of a particular product,
Association and it receives the full cooperation and support of the method, or practice. It is expected that those reproducing the material in this document for
Federal Highway Administration, United States Department of educational and not-for-profit uses will give appropriate acknowledgment of the source of
any reprinted or reproduced material. For other uses of the material, request permission
Transportation.
from CRP.
The Transportation Research Board of the National Academies was
requested by the Association to administer the research program
because of the Board’s recognized objectivity and understanding of
NOTICE
modern research practices. The Board is uniquely suited for this
purpose as it maintains an extensive committee structure from which The project that is the subject of this report was a part of the National Cooperative Highway
Research Program conducted by the Transportation Research Board with the approval of
authorities on any highway transportation subject may be drawn; it the Governing Board of the National Research Council. Such approval reflects the
possesses avenues of communications and cooperation with federal, Governing Board’s judgment that the program concerned is of national importance and
state and local governmental agencies, universities, and industry; its appropriate with respect to both the purposes and resources of the National Research
Council.
relationship to the National Research Council is an insurance of
The members of the technical committee selected to monitor this project and to review this
objectivity; it maintains a full-time research correlation staff of
report were chosen for recognized scholarly competence and with due consideration for the
specialists in highway transportation matters to bring the findings of balance of disciplines appropriate to the project. The opinions and conclusions expressed
research directly to those who are in a position to use them. or implied are those of the research agency that performed the research, and, while they have
been accepted as appropriate by the technical committee, they are not necessarily those of
The program is developed on the basis of research needs identified
the Transportation Research Board, the National Research Council, the American
by chief administrators of the highway and transportation departments Association of State Highway and Transportation Officials, or the Federal Highway
and by committees of AASHTO. Each year, specific areas of research Administration, U.S. Department of Transportation.
needs to be included in the program are proposed to the National Each report is reviewed and accepted for publication by the technical committee according
Research Council and the Board by the American Association of State to procedures established and monitored by the Transportation Research Board Executive
Committee and the Governing Board of the National Research Council.
Highway and Transportation Officials. Research projects to fulfill these
needs are defined by the Board, and qualified research agencies are The Transportation Research Board of the National Academies, the National Research
Council, the Federal Highway Administration, the American Association of State Highway
selected from those that have submitted proposals. Administration and and Transportation Officials, and the individual states participating in the National
surveillance of research contracts are the responsibilities of the National Cooperative Highway Research Program do not endorse products or manufacturers. Trade
Research Council and the Transportation Research Board. or manufacturers’ names appear herein solely because they are considered essential to the
object of this report.
The needs for highway research are many, and the National
Cooperative Highway Research Program can make significant
contributions to the solution of highway transportation problems of
mutual concern to many responsible groups. The program, however, is
intended to complement rather than to substitute for or duplicate other
highway research programs.
Published reports of the
NATIONAL COOPERATIVE HIGHWAY RESEARCH PROGRAM

are available from:
Transportation Research Board

Business Office
500 Fifth Street, NW
Washington, DC 20001
and can be ordered through the Internet at:

http://www.national-academies.org/trb/bookstore
Printed in the United States of America
COOPERATIVE RESEARCH PROGRAMS
CRP STAFF FOR NCHRP REPORT 598

Christopher W. Jenks, Director, Cooperative Research Programs
Crawford F. Jencks, Deputy Director, Cooperative Research Programs
Amir N. Hanna, Senior Program Officer
Eileen P. Delaney, Director of Publications
Hilary Freer, Senior Editor
NCHRP PROJECT 4-31

Field of Materials and Construction—Area of General Materials
Larry Lockett, Alabama DOT, Montgomery, AL (Chair)
Donald Dwyer, New York State DOT, Albany, NY
Úna Connolly, National Asphalt Pavement Association, Lanham, MD
John B. Metcalf, Louisiana State University (retired), Baton Rouge, LA
William Sheftick, Bowser-Morner, Inc., Springfield, IL
Mark B. Snyder, Bridgeville, PA
Jeff S. Uhlmeyer, Washington State DOT, Olympia, WA
John “Jack” Youtcheff, FHWA Liaison
G. P. Jayaprakash, TRB Liaison
AUTHOR ACKNOWLEDGMENTS
The research under NCHRP Project 4-31, “Tests of Recycled Aggregates for Use in Unbound Pavement
Layers,” reported herein was performed by Applied Research Associates (ARA), Inc. Dr. Athar Saeed was
the principal investigator.
Messrs. Harold Von Quintus, Jagannath Mallela, and Prithivi Kandhal assisted with the literature search
and telephone interviews. Drs. Jim W. Hall, Jr., Michael I. Hammons, and Walter Barker helped develop
the work plan. Messrs. Prithivi Kandhal, Leet Denton, and Rick Sniegowski served as consultants for all
project tasks. Laboratory tests were conducted by Boudreau Engineering, Inc., under the supervision of
Mr. Richard L. Boudreau.
Several state departments of transportation and their contractors provided recycled material for lab-
oratory testing; their contribution to this research effort is acknowledged.
FOREWORD
By Amir N. Hanna
Staff Officer
Transportation Research Board
This report contains recommendations for performance-related procedures to test and

select recycled hot-mix asphalt (HMA) and portland cement concrete (PCC) materials for
use in unbound layers of highway pavements. The report provides a comprehensive descrip-
tion of research intended to help materials engineers evaluate and select the reclaimed
asphalt pavement (RAP) and reclaimed concrete pavement (RCP) materials that should
contribute to good performing pavements. Also, the report describes procedures for the
recommended tests. The contents of this report will be of immediate interest to materials
engineers, researchers, and others concerned with the construction and performance of
asphalt and PCC pavements.
The use of RAP and RCP and materials reclaimed from structures in unbound pavement
layers should provide technical, economic, and other benefits. Although a great deal of
research has been performed on the properties of aggregates used in pavement construc-
tion, limited research has addressed the use of recycled aggregates in unbound pavement
layers. For example, research performed under NCHRP Project 4-23, and reported in
NCHRP Report 453, “Performance-Related Tests of Aggregates for Use in Unbound Pave-
ment Layers,” evaluated aggregate tests and identified a set of aggregate tests that relate to
performance of unbound pavement layers. However, the project dealt only with virgin
aggregates; it did not consider the use of recycled materials. Because RAP and RCP materi-
als are reclaimed from highway pavements, they contain binders and contaminants that are
not found in virgin aggregates. This difference in material constituents, the long exposure
of RAP and RCP materials to the elements, and constructability concerns raise questions
about the validity of the tests intended for evaluating virgin aggregates for use in evaluating
RAP and RCP materials.
Under NCHRP Project 4-31, “Tests of Recycled Aggregates for Use in Unbound Pave-
ment Layers,” Applied Research Associates, Inc. was assigned the task of recommending
procedures for performance-related testing and selection of recycled HMA and PCC mate-
rials for use in unbound pavement layers. To accomplish this objective, the researchers
reviewed relevant domestic and foreign literature; identified aggregate properties that influ-
ence the performance of pavements; identified and evaluated, in a laboratory investigation,
the aggregate tests currently used in the United States and other countries as well as poten-
tial new aggregate tests to measure performance-related properties; and recommended a set
of performance-related tests for evaluating recycled aggregates. The report documents the
work performed under NCHRP Project 4-31 and discusses the linkage between the recom-
mended tests and the performance of asphalt and concrete pavements.
The recommended set of aggregate tests can be used to evaluate and select RAP and RCP
materials for use in the unbound layers of asphalt and PCC pavements. The report includes
descriptions of those recommended test methods that are not currently being used in the
United States. These test methods will be particularly useful to highway agencies and, there-
fore, may be considered for adoption by AASHTO as standard test methods.
Appendixes A through C contained in the research agency’s final report are not published
herein. These appendixes are available on the TRB website as NCHRP Web-Only Docu-
ment 119. These appendixes discuss the following:
Appendix A: Literature Review and Background Information

Appendix B: Recommended New Aggregate Tests
Appendix C: Surface Dielectric Measurements
CONTENTS
1 Summary
2 Chapter 1 Background and Research Approach
2 Project Background
2 Research Objective
2 Scope of Study
3 Research Approach
3 Report Organization
4 Chapter 2 Pavement Performance and Recycled
Aggregate Properties
4 Background
4 Flexible Pavements
4 Rigid Pavements
4 RAP and RCP Properties
5 Mass Properties
8 Particle Properties
9 In-Service Factors Affecting RAP and RCP Performance
9 Moisture Conditions
10 State of Stress
10 Construction Method
10 Freeze-Thaw
10 Loading Rate
11 Chapter 3 Selection of Candidate Test Methods and Materials
11 Selection and Description of Test Methods
11 Screening Tests
11 Shear Strength Tests
13 Stiffness and Permeability Tests
14 Frost Susceptibility
14 Durability
14 Toughness and Abrasion Resistance
14 Selection and Description of Candidate Recycled Materials
14 RCP with Limestone
15 RCP with Gravel
16 RCP with Granite
16 RAP with Limestone
17 RAP with Granite
17 RAP with Gravel
20 Chapter 4 Laboratory Test Program and Test Results
20 Laboratory Investigation
20 Results of Laboratory Tests
20 Grain Size Analysis
20 Moisture/Density Relations
20 Static Triaxial Test
20 Repeated Load Triaxial Test Results
26 Resilient Modulus Test Results
28 Durability
35 Chapter 5 Analysis of Test Data
35 Selection of Performance-Based Test Methods
35 Durability
35 Static Triaxial Test
36 Repeated Load Triaxial Test
40 Resilient Modulus Test
45 Test Method Selection Summary
46 Selection of Recycled Materials for Intended Use
47 Selection Based on Toughness Test
48 Selection Based on Frost Susceptibility Test
48 Selection Based on Static Triaxial Test
48 Selection Based on Repeated Load Triaxial Test
48 Selection Based on Material Stiffness
51 Chapter 6 Conclusions and Suggested Research
51 Conclusions
51 Suggested Research
51 Accelerated Pavement Testing
52 In-Service Test Pavements
53 References
1
SUMMARY
Unbound pavement layers in flexible and rigid pavements generally serve to provide (1) a
working platform, (2) structural layers for the pavement system, (3) drainage layers, (4) frost-
free layers, and (5) select fill material (sometimes as part of the working platform). The prop-
erties of recycled aggregates (recycled asphalt pavement [RAP] and recycled concrete
pavement [RCP]) greatly influence their performance as unbound granular pavement layers.
Failure of an unbound pavement layer results in pavement distresses. Fatigue cracking,
rutting/corrugations, depressions, and frost heave of flexible pavements are distresses (per-
formance parameters) that can result from poor performance of aggregate in unbound base
and subbase layers. Similarly, cracking, pumping/faulting/loss of support, frost heave, and
erosion in rigid pavements can result from poor performance of subbase layers.
Factors contributing to distresses in both rigid and flexible pavements due to the poor per-
formance of unbound layers include (1) shear strength, (2) density, (3) gradation, (4) fines
content, (5) moisture level, (6) particle angularity and surface texture, (7) degradation dur-
ing construction, under repeated load and freeze-thaw cycling, and (8) drainability. Recycled
aggregate properties that were determined to affect performance of unbound pavement lay-
ers are shear strength, frost susceptibility, durability, stiffness, and toughness.
For this study, tests were conducted on RAP and RCP containing three different con-
stituent aggregates (i.e., crushed limestone, granite, and gravel) to provide a range of ma-
terials with poor to excellent performance. The recycled materials were blended with a virgin
aggregate known to provide good performance in unbound pavement layers.
Laboratory test data were analyzed. The following tests were found to produce statistically
significant performance indicators of recycled aggregates in unbound pavement layers:
• Screening tests for sieve analysis and the moisture-density relationship,
• Micro-Deval for toughness,
• Resilient modulus for stiffness,
• Static triaxial and repeated load at optimum moisture content and saturated condition
for shear strength, and
• Frost susceptibility (tube suction).
Requirements for test parameters for recycled materials were established to evaluate recy-
cled materials’ suitability for use in particular traffic and climatic conditions. The research team
developed a decision chart incorporating aggregate shear strength, stiffness, toughness, and
frost susceptibility to provide a measure of the performance potential of a particular aggregate.
The researchers also developed a validation plan to evaluate the research results in the long
term. This plan proposes accelerated pavement testing of specially constructed pavement
sections and long-term performance monitoring of in-service test pavements.
2
CHAPTER 1
Background and Research Approach
Project Background Task 2. Identify the performance parameters of pavements

that may be affected by the properties of recycled aggregates
When used appropriately, recycled materials provide good- used in unbound layers, including consideration of the layer’s
quality, cost-effective road construction materials that benefit structural behavior, constructability, and related environ-
the environment and lessen the use of raw materials. In most mental concerns.
cases, the use of recycled materials offers economic benefits,
because recycled materials often cost less than virgin aggregate Task 3. Identify and discuss the recycled aggregate properties
materials, particularly when used in reclamation jobs, and that influence the performance parameters identified in Task 2.
may also save transportation and disposal-related costs. The Chemical, mechanical, mineralogical, and physical properties
societal benefits of using recycled materials include saving shall be considered.
available natural resources and extending the life of available Task 4. Identify and evaluate—with consideration of perfor-
landfill space. Although much research has been conducted mance predictability, precision, accuracy, practicality, cost,
on using recycled materials, especially reclaimed asphalt pave- and other pertinent factors—those test procedures currently
ment (RAP) and reclaimed concrete pavement (RCP), in used for measuring the performance-related properties iden-
bound pavement layers, limited research has been reported tified in Task 3.
on the use of RAP and RCP in unbound pavement layers.
Task 5. Identify—with consideration of practicability, accu-
racy, and other relevant factors—potential new procedures
Research Objective or modifications of current test procedures for measuring those
The objective of this research was to recommend proce- performance-related properties for which no suitable test
dures for performance-related testing and selection of re- method has been identified in Task 4 and recommend proce-
cycled hot-mix asphalt (HMA) and portland cement concrete dures for further evaluation.
(PCC) materials for use as aggregates in unbound pavement Task 6. Prepare a detailed work plan for an experimental in-
layers, singularly or in combination with other materials. vestigation to evaluate and validate the most promising pro-
The research included evaluating existing aggregate tests cedures for measuring recycled aggregate properties that relate
known to predict pavement performance for their applica- to pavement performance.
bility to RAP and RCP and to develop new tests or modify
existing tests. Task 7. Execute the work plan and, based on the results of this
work, recommend sets of tests for evaluating recycled aggre-
gates used in unbound pavement layers and provide criteria
Scope of Study for interpreting test results and assessing recycled aggregate
acceptability for use in unbound pavement layers.
The research consisted of nine tasks as described below.
Task 8. Develop protocols for the tests recommended in
Task 1. Review NCHRP Report 453, and collect and review
Task 7 for which standards are not currently available, in a
other relevant domestic and foreign literature, research find-
format suitable for consideration and adoption by AASHTO.
ings, performance data, current practices, and other infor-
mation relative to the use, testing, and evaluation of recycled Task 9. Submit a final report that documents the entire re-
HMA and PCC materials in unbound pavement layers. search effort.
3
Research Approach Report Organization

The research approach included a literature search and Chapter 2 of this report discusses pavement perfor-
phone interviews with individuals representing state highway mance and recycled aggregate properties that affect pave-
agencies and relevant industry groups. NCHRP Report 453 (1) ment performance. Chapter 3 describes the methodology used
served as the initial guide for the literature search. The tele- to select candidate test methods and candidate recycled ma-
phone interviews provided information on agencies’ practices terials. Chapter 4 presents the laboratory test data. Data analy-
regarding recycling of RAP and RCP as unbound aggregate in sis is presented in Chapter 5. Chapter 6 provides research con-
base/subbase layers. The approach also included the selection clusions, recommendations, and a validation plan.
of pavement performance parameters that may be influenced Appendix A provides details of the literature search and
by the properties of recycled aggregate in unbound pavement the telephone interviews with state DOTs. Testing protocols
layers, the identification and evaluation of recycled aggregate for the recommended test methods are presented in Appen-
properties that affect pavement performance parameters, and dix B. Appendix C presents the results of surface dielectric
identification and evaluation of current aggregate test proce- measurements. Appendixes A through C are not published
dures and potential techniques that can be used to measure herein but are available as NCHRP Web-Only Document 119
relevant recycled aggregate properties. on the TRB website.
4
CHAPTER 2
Pavement Performance and Recycled

Aggregate Properties
Background the spring thaw releasing water that causes the subgrade to
weaken and pavement layers to subside or some residual dif-
Unbound pavement layers are used in flexible and rigid ferential settlement to remain.
pavements to provide one or more of the following functions:
• A working platform for construction, Rigid Pavements

• A frost blanket (frost-free layers),
•
Cracks in PCC slabs occur when the flexural strength of the
A drainage layer, and/or
• A structural layer for the pavement system. concrete is exceeded by the imposed stresses. PCC pavements
usually show fatigue failure due to repeated applications of
Fatigue cracking, rutting/corrugations, depressions, and stresses less than the flexural strength. Fatigue cracking occurs
frost heave of flexible pavements can be attributed, at least in in the form of longitudinal cracks and corner breaks attributed
part, to poor performance of granular base and subbase layers. to poor unbound material performance. Load associated lon-
Cracking, pumping/faulting/loss of support, frost heave, and gitudinal cracks along the wheel paths are due to a combina-
erosion of rigid pavements can also be attributed to poor per- tion of factors including wheel load and thermal stresses and
formance of granular base and subbase layers. These distresses moisture variations. Corner breaks are usually caused by a loss
and granular base/subbase contributing factors are described of support due to pumping of unbound materials and/or
in Table 2.1 for flexible pavements and Table 2.2 for rigid reduced strength from increased moisture content.
pavements. Pumping in rigid pavements usually results in erosion
and migration of the unbound material from underneath the
concrete slabs leading to a gradual reduction in slab support
Flexible Pavements and slab faulting in extreme cases. Frost heave can also cause
Flexible pavements with thin and thick HMA surfaces over shear strength loss in unbound layers resulting in pavement
unbound aggregate bases may exhibit different failure modes. roughness and faulting. Frost action, with slow freezing and
Flexible pavements with thick HMA surfaces usually exhibit available water, results in the formation of ice lenses which
rutting failure while thin HMA surfaced flexible pavements release large amounts of water during spring thaw, weakening
often exhibit fatigue cracking. the unbound layer/subgrade and leading to cracking, pumping,
Depressions are localized pavement surface areas with ele- and faulting.
vations slightly lower than those of the surrounding pavement.
Depressions are generally the result of localized areas in the RAP and RCP Properties
base or subgrade caused by low initial density that have further
compacted under traffic load repetitions or depressed into the The performance of pavements built with unbound base
weakened subgrade. and subbase layers incorporating recycled aggregates can be
Frost heave causes differential surface movement, which affected by physical, chemical, and mechanical properties of
results in cracking and uneven surface conditions. Thick ice the recycled aggregate particles and the proportion in which
lenses are formed due to slow freezing during winter causing they are mixed with virgin aggregate (zero to 100 percent).
the pavement surface to distort. These ice lenses melt during Some of the properties are listed in Table 2.3.
5
Table 2.1. Flexible pavement performance parameters and contributing factors.
Distress Description of distress Unbound layer failure mechanism Contributing factors

Fatigue cracking first appears as Lack of base stiffness causes high Low modulus of the base
fine, longitudinal hairline cracks deflection/strain in the HMA surface layer
running parallel to one another in under repeated wheel loads, Low density of the base
the wheel path and in the direction resulting in fatigue cracking of the layer
Fatigue Cracking
of traffic; as the distress progresses HMA surface. High flexibility in Improper gradation
the cracks interconnect, forming the base allows excessive bending High fines content
many-sided, sharp angled pieces; strains in the HMA surface. The High moisture level
eventually cracks become wider same result can also be due to Lack of adequate particle
and, in later stages, some spalling inadequate base thickness. Changes angularity and surface
occurs with loose pieces prevalent. in base properties (e.g., moisture texture
Fatigue cracking occurs only in induced) with time can render the Degradation under
areas subjected to repeated base inadequate to support loads. repeated loads and
loadings. freeze-thaw cycling
Rutting appears as a longitudinal Inadequate shear strength in the base Low shear strength
surface depression in the wheel allows lateral displacement of Low base material density
path and may not be noticeable particles with applications of wheel Improper gradation
Rutting/Corrugation
except during and following loads and results in a decrease in the High fines content
rains. Pavement uplift may occur base layer thickness in the wheel High moisture level
along the sides of the rut. Rutting path. Rutting may also result from Lack of particle angularity
results from a permanent densification of the base due to and surface texture
deformation in one or more inadequate initial density. Changes Degradation under
pavement layers or subgrade, in base (mainly degradation repeated loads and
usually caused by consolidation producing fines) can result in freeze-thaw cycling
and/or lateral movement of the rutting. The base can also lose shear High moisture content
materials due to load. strength from moisture-induced coupled with traffic can
damage, which will cause rutting. contribute to stripping
Depressions are localized low Inadequate initial compaction or Low density of base
areas in the pavement surface nonuniform material conditions result material
caused by settlement of the in additional reduction in volume with Low shear strength of the
Depressions
foundation soil or consolidation load applications. Changes in base material combined

in the subgrade or base/subbase material conditions due to poor with inadequate surface
layers due to improper durability or frost effects may also thickness
compaction. Depressions result in localized densification with
contribute to roughness and cause eventual fatigue failure.
hydroplaning when filled with
water.
Frost heave appears as an upward Ice lenses are created within the Freezing temperatures
bulge in the pavement surface base/subbase during freezing Source of water
and may be accompanied by temperatures, particularly when Permeability of material
surface cracking, including freezing occurs slowly, as moisture high enough to allow free
alligator cracking with resulting is pulled from below by capillary moisture movement to the
Frost Heave
potholes. Freezing of underlying action. During spring thaw large freezing zone, but low
layers resulting in an increased quantities of water are released from enough to also allow
volume of material cause the the frozen zone, which can include suction or capillary action
upheaval. An advanced stage of all unbound materials. to occur
the distortion mode of distress
resulting from differential heave
is surface cracking with random
orientation and spacing.
Properties of recycled aggregate that are considered relevant to prevent pumping and faulting. However, adequate shear
to its use in unbound pavement layers are listed in Table 2.4. strength is also needed for construction purposes and to pro-
Table 2.5 shows the links between performance parameters vide protection from base shear under pavement joints (1).
and laboratory test measures. Aggregate mass properties are affected by aggregate particle
properties, particularly gradation, shape, texture and angu-
larity, freeze-thaw durability, and toughness.
Mass Properties
The properties listed in Table 2.4 are properties of aggregate
Shear Strength
mass (mass properties) that describe the behavior of the aggre-
gate layer as a continuum. In flexible pavements, shear strength Aggregate shear strength has been identified as the single
is the most important property, although stiffness is also im- most important aggregate mass property of unbound pave-
portant (shear strength is very closely related to stiffness). For ment layers for both flexible and rigid pavements (1). Shear
rigid pavements, permeability is an important mass property strength of unbound pavement layers is typically measured in
6
Table 2.2. Rigid pavement performance parameters and contributing factors.
Distress Description of distress Unbound layer failure mechanism Contributing factors

Low base stiffness and
Cracks transverse to the pavement Inadequate support or loss of
shear strength
centerline, generally within the support from the unbound aggregate
Pumping of base/subgrade
center one-third of the slab. base/subbase resulting from
fines causing loss of
Corner breaks and diagonal cracks settlement or erosion can increase
support
appear as top down hairline cracks tensile stresses in the slab under
Low density in base
across slab corners where the repeated wheel loads and result in
Improper gradation
crack intersectsthe joints less increased transverse or corner or
High fines content
than 6 ft from the corner; cracking diagonal cracking which initiates at
High moisture level
progresses to result in several the bottom or top of the slab (corner
Lack of adequate particle
broken pieces with spalling of cracks at the top). When a crack
angularity and surface
Cracking
crack and faulting at the crack or develops, increased load is placed

texture
joint up to ½ in. or more. on the base, resulting in deformation
Degradation under
The corner break is a crack within the base and surface
repeated loads and
completely through the slab (as roughness of the pavement; the
freeze-thaw cycling
opposed to corner spalls, which crack introduces moisture to the
intersect the joint at an angle). base, resulting in further loss of
support and possibly erosion and
faulting thereby resulting in further
deformation and roughness. Corner
breaks (and associated faulting) are
caused by lack of base support from
erosion and pumping of the base
material and freeze-thaw softening
of the base.
Pumping and faulting begin as Pumping involves the formation of Poor drainability (low
water seeping or bleeding to the a slurry of fines from a saturated permeability)
Pumping/Faulting
surface at joints or cracks and erodible base or subgrade, which is Free water in base
progress to fine material being ejected through joints or cracks in Low base stiffness and
pumped to the surface; the the pavement under the action of shear strength
ultimate condition is an elevation repetitive wheel loads. Highly erodible base
differential at the joint termed High fines content
faulting. Pumping action is Degradation under
caused by repeated load repeated loads and
applications that progressively settlement
eject particles of unbound
material from beneath the slabs.
Differential heave during freezing Ice lenses are created within the Freezing temperatures
and formation of ice lenses causes base/subbase during freezing Capillary source of water
roughness due to uneven temperatures, particularly when Permeability of material
Frost Heave
displacement of PCC slabs; thaw freezing occurs slowly, as moisture high enough to allow free
weakening results in loss of is pulled from below by capillary moisture movement to the
support from base and subgrade action and migrates toward the freezing zone
which may cause pumping and freezing front. During spring thaw,
faulting and corner breaks; under large quantities of water are
heavy loads, the loss of support released from the frozen zone,
can result in cracking of slabs. which can include all unbound
materials.
Table 2.3. Recycled aggregate particle properties that

influence pavement performance.
Physical properties Chemical properties Mechanical properties

Particle gradation and shape Solubility Particle strength
(max/min sizes) Base exchange Particle stiffness
Particle surface texture Surface charge Wear resistance
Pore structure, absorption, Chemical reactivity Resistance to degradation
porosity (resistance to attack by Particle shape of abraded
Permeability (hydraulic chemicals, chemical fragments
properties) compound reactivity,
Specific gravity oxidation and hydration
Thermal properties reactivity, organic
Volume change (in wetting material reactivity)
& drying) Chloride content
Freezing/thawing resistance pH-level
Deleterious substances
7
Table 2.4. Relevance of recycled material mass properties for various

applications.
Relevance of Mass Property to the Use of Recycled Material as

Mass Property of
Material Structural Construction Drainage Frost Control
Select Fill
Layer Platform Layer Blanket Pumping
Shear Strength Y Y N N N N
California Bearing
Y Y N N N Y
Ratio (CBR)
Cohesion & Angle
Y N N N N N
of Internal Friction
Resilient or Com-
Y Y Y Y Y Y
pressive Modulus
Density Y Y N Y Y Y
Permeability N N Y Y Y N
Frost Resistance Y N Y Y N Y
Durability Index Y N Y Y Y N
Resistance to
Y N N N N N
moisture damage
Y: Relevant; N: Not relevant
the “drained” condition (pore pressures allowed to dissipate content (2). Of these, construction and in situ conditions dic-
during testing). The Shear strength, s, is mainly a function of tate moisture and density. It is important that the measures of
angle of internal friction, φ, and to a smaller degree its cohesion, internal friction angle and cohesion reflect the conditions that
c, and may be described by the Mohr-Coulomb equation: are likely to occur during the life of the pavement. Important
aspects of the state-of-stress are the stress magnitude, number
s = c + σ tan φ of stress repetitions, and rates of loading. In general, static test-
ing procedures are not appropriate for characterizing the be-
in which σ is the normal effective stress on the failure plane. havior of aggregate materials subjected to the impulse type
Factors influencing the shear strength of an aggregate include repeated loading caused by moving wheel loads. If the pavement
gradation, density, plasticity index, particle geometric charac- is subject to water infiltration, the measure of shear strength
teristics (shape, angularity, surface texture), and moisture must reflect severe moisture conditions. If the aggregate layer is
Table 2.5. Links between aggregate properties and performance.
Pavement Performance Related aggregate

Test measures
type parameter property
Resilient modulus, Poisson’s ratio, gradation, fines content,
Fatigue Cracking Stiffness particle angularity and surface texture, frost susceptibility
degradation of particles, density
Failure stress, angle of internal friction, cohesion, gradation,
Rutting,
Shear Strength fines content, particle geometrics (texture, shape,
Flexible
Corrugations
angularity), density, moisture effects
Particle strength, particle degradation, particle size,
Toughness
gradation, high fines
Fatigue Cracking,
Durability Particle deterioration, strength loss
Rutting,
Corrugations Frost Permeability, gradation, percent minus 0.02 mm size,
Susceptibility density, nature of fines
Permeability Gradation, fines content, density
Failure stress, angle of internal friction, cohesion, gradation,
Shear Strength fines content, particle geometrics (texture, shape,
angularity), density, moisture effects
Cracking,
Stiffness Resilient modulus, Poisson’s ratio
Pumping, Faulting
Rigid
Toughness Particle strength, particle degradation, gradation

Durability Particle deterioration, strength loss
Permeability Gradation, fines content, density
Cracking,
Permeability, gradation, percent minus 0.02 mm size,
Pumping, Faulting, Frost Susceptibility
density, nature of fines
Roughness
8
subject to freezing and the material is even moderately frost shear strength and stiffness to decrease. The increased pore
susceptible, then the measure of shear strength must reflect pressures, along with the lubricating effect of the retained
strength loss due to freeze-thaw. water, could lead to rapid degradation of the aggregate.
Resistance to permanent deformation, which is nearly syn-
onymous with shear strength, is an important characteristic
Frost Susceptibility
for aggregates that are to be used as base course materials in pave-
ments. The Aggregate Handbook states that, although consid- Frost susceptibility refers to the degree to which a soil mass
erable emphasis is being placed on resilient moduli, permanent is affected by the action of freeze-thaw in the presence of water.
deformation characteristics are often more important from a Frost action occurs in frost-susceptible soils when freezing tem-
practical viewpoint (2). This property refers to an aggregate’s perature penetrates into the pavement structure and source
ability to resist permanent deformation under repeated loads, of water exists.
which is quantified by repeated load tests. Frost heave is usually not uniform; generally differential
heaving occurs, causing surface irregularities, roughness, and
possible cracking. Non-frost-susceptible aggregate layers
Stiffness
are used in pavements to limit the frost from reaching frost-
The stiffness of the aggregate layer in flexible pavements is susceptible soils and/or by reducing the volume of frost-
important, because the aggregate layer is an integral part of susceptible subgrades subjected to freezing temperatures.
the pavement structural system. Because aggregate properties However, the shear strength and stiffness of the material is
that influence shear strength also influence stiffness, results greatly reduced in the spring when frost-susceptible materials
of shear strength and stiffness tests are expected to be highly thaw.
correlated. The shear strength test procedure can usually be The frost susceptibility of an aggregate mass depends pri-
adapted to also measure the resilient modulus. marily on its permeability. Frost-susceptible materials have
permeability that permits the movement of capillary water
from a water source to the freeze front such that ice lenses are
Density
formed. If the permeability is sufficiently low, moisture can
Density of unbound aggregate layers typically refers to the move at a rate to form ice of a magnitude to be detrimental.
bulk dry unit weight of a compacted mass of aggregate particles. Fines content defines the permeability and thus the frost sus-
Insufficient density will cause depressions and rutting due to ceptibility of an aggregate mass.
densification. Other mass properties are greatly affected by the
mass density and, therefore, mass density must be considered
during the measurement of the other mass properties. In- Particle Properties
creasing density results in increased shear strength and in- Particle Size Distribution
creased stiffness, but it may lead to a reduction in permeability.
Particle size distribution or gradation is a measure of the
relative size distributions of different particles in the aggregate
Permeability mass and is an important indicator of field performance (1).
Permeability refers to the ability of the aggregate layer to Aggregate gradation can be dense graded to provide high shear
allow water to flow through it; the rate of flow is usually differ- strength and stiffness or open-graded and free draining to re-
ent in the horizontal direction than in the vertical direction. duce damage due to excessive moisture and frost action. The
The quantity of flow increases as the coefficient of perme- amount of material passing the No. 200 sieve and the nature of
ability increases. When an aggregate layer is used in a rigid that material are usually controlled to limit frost susceptibility
pavement to prevent pumping, the permeability of the ag- and to provide sufficient permeability. Particle size distribution
gregate mass is the primary mass property that dictates the also affects constructability.
performance of the aggregate layer. Aggregate properties that Measuring gradation is influenced by particle shape and tex-
influence the coefficient of permeability include gradation, ture (3). Rod-shaped elongated particles may not pass through
compacted density (including void ratio and porosity), and square sieve openings because of their orientation during a
degree of saturation. Changing the aggregate particle properties sieve test, making gradation appear coarser.
to increase permeability can affect shear strength and stiffness
adversely. Permeability also influences the ratio of dry to wet Particle Shape, Texture, and Angularity
shear strength and stiffness. Unbound pavement layers with
low permeability values usually retain water and may, under Lees (4) defines the shape of aggregate particles as cubical,
repeated dynamic loads, develop pore pressures that cause equi-dimensional, blade, disk, or rod-shaped. Angular-shaped
9
particles provide higher internal friction and shear strength; ment distress (1). The sulfate soundness test does not evalu-
thus, it is desired that a certain percentage of particles be ate environmental effects on RAP and RCP well and is usu-
crushed. The mineral and geological properties of the rock ally waived with the assumption that the aggregate used in
formation and the crushing process define the shape of the the original HMA and PCC pavement had been tested for
crushed particles. durability and had to meet the durability specifications (6).
Surface roughness and irregularities are termed as particle
texture. Contrasts of texture are a smooth river gravel with
Plasticity of Fines Fraction
polished surface as compared to a crushed limestone or gran-
ite with harsh surfaces. Thompson (5) has shown that the tex- Plasticity of fines fraction may indicate the effect of mois-
ture of both coarse and fine aggregate particles is important ture on aggregate performance. Fines with high plasticity
to achieve field shear strength. Current practice often does tend to attract and retain greater quantities of water and cause
not consider particle texture directly. The Index of Aggregate greater loss of shear strength and stiffness than will fines of
Particle Shape and Texture (ASTM D 3398) considers parti- low plasticity. Because the fines produced during processing
cle texture indirectly. RAP and RCP are relatively low and are non-plastic, the in-
fluence of the plasticity of the fines on the performance is low
but the magnitude of the fines is important.
Toughness and Abrasion Resistance
An aggregate’s ability to resist mechanical degradation
In-Service Factors Affecting
during hauling and construction operations is termed tough-
RAP and RCP Performance
ness, and abrasion resistance is a measure of toughness. Impact,
repeated stresses, and continuous abrasion cause mechanical In-service performance of RAP and RCP as aggregate in un-
degradation. Aggregates are subjected to impact stresses during bound pavement layers is influenced by moisture conditions,
handling, processing, and compaction and repeated stresses state of stress, processing and construction method, loading
during the service life of a pavement. Aggregate stresses, which rate, and freeze thaw—factors that generally also have similar
are transmitted through particle contacts, are highest when effects on virgin aggregate material. Particle size gradation,
they are supporting the loads of passing vehicles. quality of original aggregate, production process, and binder
Mechanical degradation of aggregate with low toughness content may also affect their performance as unbound pave-
causes aggregate gradation changes and a decrease in perme- ment layers.
ability, decrease in wet shear strength, or an increase in mass
density. An increase in mass density of very open-graded ma-
Moisture Conditions
terial can lead to rutting in flexible pavements or faulting in
rigid pavements. The LA abrasion test is a commonly used The adverse effects of water in the pavement structure can be
method for obtaining a relative measure of mechanical tough- reduced by improving drainage and by selecting aggregate ma-
ness. Other testing options include repeated triaxial tests and terials that are least affected by the presence of water. For good
cyclic loading by a gyratory test machine. performance, aggregate must possess adequate shear strength
and stiffness when wet and subjected to repeated loadings. This
can be achieved by controlling the amount of material passing
Durability and Soundness
the No. 200 sieve to less than about 8 percent (1). This aspect
An aggregate’s ability to resist degradation due to environ- is not a concern for RAP and RCP because of the relatively
mental or chemical effects is measured in terms of its dura- small amount of fines produced during processing. RAP is
bility and soundness. Chemical attack of unbound aggregate also susceptible to the presence of water in that the asphalt
bases is very unusual and was not considered in this study. The coating may start to strip from the aggregate, raising perme-
aggregate particles’ resistance to the cumulative effects of cyclic ability problems.
wetting and drying and freeze-thaw is a common concern. Liu et al. (7) and Liu and Lytton (8) stated that a drainage
Sulfate soundness tests are often used to obtain a relative mea- time of 5 hours to reach 85 percent saturation from 100 per-
sure of an aggregate’s durability. Although the particle degra- cent saturation is acceptable for a base material, drainage
dation during this test is caused by crystallization pressures times in excess of 10 hours were unacceptable, and those be-
from magnesium or sodium sulfate, it provides a measure of tween 5 and 10 hours were marginal. Thompson (9) noted
resistance to damage caused by wet-dry or freeze-thaw cycles. that the quantity and nature of the fines fraction directly in-
Degradation of the base material and the changes in its related fluence moisture sensitivity and permeability. A base course
properties (i.e., stiffness and shear strength) can lead to pave- with high fines is likely to have low permeability, especially if
10
the fines consist of clay-sized particles. The use of more open- because the asphalt coating on RAP particles acts as a stress ab-
graded aggregates for pavement base courses can decrease sorber and hardened cement paste on RCP particles provide
moisture sensitivity of RAP material (less fines, more film additional degradation/abrasion resistance. The increase in
thickness) and thus the probability of stripping of asphalt be- fines content for RAP (0.60 percent) and RCP (1.6 percent)
cause of aggregates’ ability to limit high pore pressures. was much less than that of virgin aggregate (3.6 percent) (11).
Other aspects of construction, such as crushing, handling,
and stockpiling of RAP and RCP can significantly alter the
State of Stress
mass strength, stiffness, permeability, or frost susceptibility.
The performance of unbound aggregate must be judged at It is not practical to use equipment and methods that simulate
the state-of-stress representative of field conditions (2). In construction practice in the laboratory for routine aggregate
flexible pavements, an unbound aggregate layer may be used evaluation. Therefore, the aggregate tested for mass properties
as a structural layer. If a base or a subbase is used as a struc- should match the final aggregate in the pavement.
tural layer, the unbound aggregates are subjected to high
vertical and shear stresses. In rigid pavements, the unbound
Freeze-Thaw
aggregate layer is often placed directly under the PCC slab.
Unbound aggregate layers used in rigid pavements and as Freezing and thawing can be extremely detrimental to ag-
subbases in flexible pavement are subjected to large confin- gregate mass properties. However, the performance of RAP and
ing stress but relatively low shear stress; the most severe stress RCP as aggregate in unbound layers in a freeze-thaw environ-
conditions occur during construction. ment will depend on in-service conditions and the number of
Repeated load triaxial tests used to characterize the response freeze-thaw cycles. Another in-service factor is the uniformity
of granular materials have shown that open-graded granular of the aggregate mass. Uniform layers may heave at a uniform
materials with larger maximum sizes typically display a “stiffer” rate and, thus, may not present a functional problem for the
response than dense-graded materials with a smaller maxi- pavement. Reduced mass density due to frost heave can cause
mum size as indicated by the resilient response under repeated reduced shear strength. Frost susceptibility tests that simulate
loads (1). During repeated-load testing, granular materials the number of freeze-thaw cycles, rate of freezing, availability
exhibit a plastic strain component (or permanent deforma- of water, and degree of drainage have been developed. These
tion). The permanent deformation response of granular tests are complex and difficult to run and are not suited for
material tested under fixed conditions of density and moisture routine aggregate classification (1).
depends on the magnitude of the repeated stress state. Both
the maximum principal stress, σ1, and the stress ratio, σ1/σ3,
Loading Rate
influence the permanent deformation behavior of granular
materials. For a given confining stress (σ3), permanent de- Aggregate layers in pavements are subject to many cycles
formations will increase for higher values of σ1. For a given of moving wheel loads. The rate of loading varies from static
stress ratio, permanent deformations will increase with the and slow-moving loads experienced in parking lots, at stop-
corresponding increases in both σ1 and σ3 (1). lights, and at highway intersections, to fast rates of loading cor-
responding to interstate highway speeds. Increased loads may
also be experienced due to wheel impacts at bumps or corru-
Construction Method
gations. Much work has been done to estimate the loading
Pavement structures, particularly flexible pavements de- rates that correspond to highway traffic loading for use in
signed for heavier wheel loads and larger numbers of load strength and stiffness tests. Also test procedures have been
applications, typically use high density in the aggregate mass. developed for measuring stiffness, both in the laboratory and
Because more compactive effort is required to achieve these in the field, when the material has reached a steady-state con-
higher densities, the aggregate is subjected to higher stresses dition under cyclic loading. For granular material and nor-
during construction that can result in aggregate degradation. mal highway speeds, the rate of loading has been shown to
However, Chini and Kuo (10) stated that, irrespective of its have insignificant influence on the strength and stiffness
strength, crushed concrete does not break down during han- properties of the aggregate mass. The stiffness of unbound
dling and compaction. base materials with high amounts of RAP can be affected by
A comparison of RAP and RCP degradation due to com- temperature and frequency or loading rate. Several re-
paction to that of virgin aggregate material determined that searchers have found that a steady-state, similar to field con-
RAP and RCP do not degrade during construction, possibly ditions, occurs after relatively few cycles of loading (1).
11
CHAPTER 3
Selection of Candidate Test Methods

and Materials
Selection and Description mance and is used by most agencies in aggregate selection.
of Test Methods The gradation is an indicator of permeability, frost suscepti-
bility, and shear strength.
Laboratory tests are used to characterize aggregates as a Laboratory compaction is used to determine the antici-
construction material, to ensure specification compliance, and pated density achievable in the field and for fabrication of
to evaluate the strength and durability properties. A number of laboratory specimens for other tests. Compaction of aggre-
laboratory tests have been developed, mostly along empirical gate materials generally results in increasing density, shear
lines, to estimate performance and to identify potentially poor strength, and stiffness and decreasing permeability. Density
performers. The proportions and properties of RAP and RCP increases with increasing moisture content to a point of
in the unbound pavement layers define the performance of maximum density at the optimum moisture content (OMC),
the unbound pavement layer. Due to the particulate nature of beyond which density decreases. The OMC is a function of
unbound aggregate layers, their mechanical properties also compactive effort.
depend on stress state and environmental conditions. For pave-
ment applications, tests have been developed to measure four
categories of aggregate properties and characteristics: (1) stiff- Shear Strength Tests
ness or modulus, (2) shear strength, (3) permanent deforma-
tion, and (4) durability. Shear strength was identified in NCHRP Report 453 (1) as
This section discusses the test methods selected to deter- the single most important property governing unbound pave-
mine the performance-related properties of RAP and RCP. ment layer performance. The shear strength tests selected for
Performance predictability, precision, accuracy, practicality, the laboratory investigation were the static triaxial shear tests
and cost were considered. Laboratory tests recommended in (AASHTO T 296) and the repeated load triaxial test recom-
NCHRP Report 453 (1) were re-evaluated for applicability to mended by NCHRP Report 453.
RAP and RCP; results are shown in Table 3.1. The static triaxial test is simple to conduct and is well ac-
Selected tests for laboratory investigation are shown in cepted in geotechnical applications. The test is conducted on
Table 3.2. These tests had a high composite rating in the eval- specimens compacted to 95 percent of the maximum dry den-
uation shown in Table 3.1. The selected tests can be used to sity at the OMC as determined by AASHTO T 180.
evaluate factors that influence the performance of recycled ag- Repeated load triaxial tests are conducted on triplicate
gregate and differentiate between good and poor performance samples prepared at 95 percent of the maximum dry den-
potential. The selected test methods could be performed by sity at OMC as determined by AASHTO T 180 and on sat-
most state DOTs at a reasonable cost. urated samples (prepared at OMC) using a closed-loop
The tests listed in Table 3.2 are discussed in the following servo-hydraulic test system. The test provides a relative
section; additional detail is provided elsewhere (1, 2). measure of an aggregate’s ability to resist permanent defor-
mation. The repeated load tests are conducted at a confin-
ing pressure of 15 psi. An array of load increments is applied,
Screening Tests
with 1,000 repetitions at each load level (Table 3.3). The load
Screening tests included sieve analysis (AASHTO T 27 and level is increased until the aggregate sample fails in shear or
T 11) and moisture-density relations (AASHTO T 180). the permanent deformation reaches 10 percent. The test time
Aggregate mass gradation is an indicator of aggregate perfor- depends on the selected number of load cycles per load level,
12
Table 3.1. Rating of potential test methods for evaluating recycled aggregates.
Property Performance Practi- Com-

Test Accuracy Precision Cost Composite
measured predictability cality plexity
Static Triaxial Shear F G H FS G M H
Repeated Load Triaxial G G H C G M H
Texas Triaxial F G M FS F M M
Illinois Rapid Shear F-G G M FS G M M–H
Confined Compression F F M S F L M
Direct Shear F F L FS F M M
Shear Strength Gyratory Shear F F M C F M M
k-Mould G G M C F M M
CBR F F M S F L M
Hveem Stabilometer F F M S F L M
Hollow Cylinder G G L VC L H L
Dynamic Cone Penetrometer F F M S F L M
Lab Rut-Tester G F L C F H M
Resilient Modulus G G H C G M H
Stiffness Var. Conf. Pres. Modulus F F L VC F H L
Resonant Column P P L C P M L
Frost Susceptibility Test F F L C P H L
Frost
Tube Suction Test G G M FS G M H
Susceptibility
Index Tests F G H S F L H
Constant Head F F M FS F L M
Permeability Falling Head F F H FS F L M
Pressure Chamber F F H FS F M M
Horizontal Permeameter F F H FS G M M
LA Abrasion F F M S F L M
Aggregate Impact Value F F F S F L M
Aggregate Crushing Value F F F S F L M
Toughness Aggregate Abrasion Value P P P FS P L L
Micro-Deval G F M S F L H
Durability Mill P P P FS P L L
Gyratory Test P P P FS F M L
Tube Suction Test G G M FS G M H
Sulfate Soundness P P P F F L L
Durability Freezing and Thawing P P P FS F M L
Canadian Freeze-Thaw G G M FS F L H
Aggregate Durability Index F F H FS F L M
Unconfined Freeze Thaw F F H FS F M M
Shape/ Surface Texture Index F F M S F L M
Flat and Elongated Particles P P L C P L L
Particle
Percent Fractured Particles P P L C P L L
Geometric
Properties Uncompacted Void Content P P L C P L L
Digital Image Analysis P P L C F H L
Atterberg Limits F F M S F L M
Rating Scale:
Performance Predictability - G = good, F = fair, P = poor
Accuracy - G = good, F = fair, P = poor
Practicality - H = high, M = medium, L = low, F = fair, P = poor
Complexity Levels - S = simple, FS = fairly simple, C = complex, VC = very complex
Precision - G = good, F = fair, P = poor, L = low
Cost - H = high, M = medium, L = low
Composite – H = high, M = medium, L = low (based on relative ratings of other factors)
Notes: 1. All ratings are average subjective evaluations of research team.
2. The composite rating is based on the relative ratings for each category.
Table 3.2. Tests selected for the laboratory test program.
Aggregate
Test method Test reference Test parameter
property
Sieve Analysis AASHTO T 27 and T 11 Particle size distribution
Screening Tests Maximum dry density
Moisture-Density Relationship AASHTO T 180 and optimum moisture
content
Static Triaxial Shear AASHTO T 296 c, φ, shear strength
Shear Strength
Repeated Load Triaxial NCHRP Report 453 (1)
Permeability Saturated Repeated Load Triaxial NCHRP Report 453 (1)
Stiffness Resilient Modulus NCHRP Report 453 (1)
Frost Tube Suction Test NCHRP Report 453 (1)
Susceptibility Index Method U.S. Army Corps of Engineers, F categories
Toughness Tests Micro-Deval AASHTO TP 58
Durability Canadian Freeze-Thaw MTO LS-614
13
Table 3.3. Stress control for repeated load triaxial test.
Confining Contact Stress Cyclic Stress

Sequence No. No. of Cycles
Pressure (psi) (psi) (psi)
PC 15 1 10 50
1 15 1 10 1000
2 15 1 20 1000
3 15 1 40 1000
4 15 1 60 1000
5 15 1 80 1000
6 15 1 100 1000
7 15 1 120 1000
8 15 1 140 1000
9 15 1 160 1000
10 15 1 180 * 1000
* A 5,000-pound load cell can accommodate a load equivalent to about 180 psi of axial stress on a
nominal 6-inch diameter test specimen (approximately 5,110 pounds).
the load cycle rate, and the number of load levels. The num- cycles at each load increment are considered conditioning for
ber of load cycles per load level and the load cycle rate can that load increment.
be set, but the number of load levels depends on the strength
of the aggregate sample. To keep the testing time reason-
Stiffness and Permeability Tests
able, 1,000 cycles at each load level are applied at a rate of
60 cycles per minute (Table 3.3), thus requiring approxi- The repeated load triaxial tests are conducted to obtain re-
mately 17 minutes to complete each load level. silient modulus (MR), a measure of stiffness. Data from cycles
A haversine load pulse of 0.1-sec load duration, shown in 96-100 of each 1,000-cycle loading level are used to compute
Figure 3.1, is used to apply load to the test specimen. Each load the MR (per the procedure described in AASHTO T 307-99).
pulse is followed by a 0.9-sec relaxation period. This procedure This allows 95 cycles of conditioning prior to measuring
does not allow for sample conditioning, but the first few load strain response to loading. For determining design param-
0.8
0.6
Load Factor
0.4
0.2
0
0 0.02 0.04 0.06 0.08 0.1
Time (s)
Figure 3.1. Load pulse for repeated load triaxial tests.

14
eters, conducting a full MR test to determine the stiffness as a and thawed at room temperature for 8 hours. The material is
function of the state-of-stress is not necessary. The saturation then drained, dried, and re-sieved using the original sieve sizes.
phase of the saturated repeated load triaxial tests provides a The weighted average loss for the sample is then determined
measure of material permeability. from the original grading and the percent loss from all three
fractions.
Frost Susceptibility
Frost susceptibility of the materials is determined using the
tube suction test (Texas Method 144 E), which determines the The aggregate toughness and abrasion resistance is deter-
moisture retention potential of an aggregate based on the sur- mined using the Micro-Deval test (AASHTO TP 58-00).
face dielectric values of a compacted specimen after a 10-day The test is performed on an aggregate sample consisting
capillary soak in the laboratory. For materials with high suc- of 750 grams of 3⁄4- to 1⁄2-inch (19- to 13-mm) material and
tion potential and sufficient permeability, substantial amounts 750 grams of 1⁄2- to 3⁄8-inch (13- to 9.5-mm) material. The
of unbound water rise within the aggregate matrix during sample is soaked in water for 24 hours and then placed in a jar
soaking and lead to higher dielectric values in the test. Con- mill with 2.5 liters of water and an abrasive charge consisting
versely, non-frost-susceptible materials allow little moisture to of 11 pounds (5 kg) of 3⁄8-inch-diameter (9.5-mm) steel balls.
reach the surface and have lower dielectric values at the end of The jar, aggregate, water, and abrasive charge are revolved at
testing. The tube suction test procedure requires preparing two 100 rpm for 2 hours. The sample is then washed and dried. The
samples compacted at OMC. On the first sample, a triaxial amount of material passing the No. 16 sieve is determined, and
strength is measured at OMC. On the second sample, a full the loss, expressed as a percent by weight of the original sam-
10-day tube suction test is run, after which the sample is ple, is calculated.
subjected to triaxial strength testing. The change in triaxial
strengths provides a measure of the material’s loss in strength
Selection and Description of
after exposure to moisture.
Candidate Recycled Materials
Tests were conducted on RAP and RCP containing three
Durability
different constituent aggregates (crushed limestone, granite,
The Canadian Freeze-Thaw test (MTO LS-614) was selected and gravel) to provide a range of performance, as shown in
to evaluate recycled aggregate durability. The test assesses ag- Table 3.4. The recycled materials were blended with a virgin
gregate durability by cyclic freezing and thawing in the pres- aggregate that is known to provide good performance when
ence of moisture. The test is conducted by placing three frac- used in unbound pavement layers.
tions of aggregate into separate 1-liter jars (3⁄4 inch to 1⁄2 inch
[1,250 grams], 1⁄2 inch to 3⁄8 inch [1,000 grams], and 3⁄8 inch to
RCP with Limestone
No. 4 [500 grams]). Aggregate samples are soaked for 24 hours
in a 3-percent NaCl solution before application of freeze and RCP with limestone (RCP-LS-IL) was obtained from a sec-
thaw cycles. After the 24-hour soak period, the samples are tion of Dan Ryan Expressway (Interstate 94/90) that extends
drained, sealed and cycled 5 times, frozen for 16 hours at 0°F, from downtown Chicago through south Chicago. This section
Table 3.4. Selected materials and expected performance potential.
Expected Performance
Proposed Materials
Potential
100 % RCP (granite)
Excellent
Limestone aggregate (virgin - for blending)
100 % RCP (limestone) 50 % RCP (granite) + 50 % limestone aggregate
Very Good
50 % RCP (limestone) + 50 % limestone aggregate
100 % RCP (gravel) 50 % RAP (limestone) + 50 % limestone aggregate
Good
100 % RAP (limestone) 50 % RAP (gravel) + 50 % limestone aggregate
50 % RAP (granite) + 50 % limestone aggregate
Fair
100 % RAP (granite)
50 % RAP (gravel or soft limestone) + 50 % limestone aggregate
Poor
100 % RAP (gravel or soft limestone)
15
Table 3.5. Gradation for Illinois DOT CA 6 coarse

aggregate.
Sieve Size Percent passing (percent finer)

1.5 in (37.5 mm) 100
1.0 in (25 mm) 95 5
3/4 in (19 mm) -- b
1/2 in (12.5 mm) 75 15
3/8 in (9.5 mm) -- b
No. 4 (4.75 mm) 43 13
No. 8 (2.36 mm) -- a
No. 10 (2.00 mm) -- a
No.16 (1.18 mm) 25 15
No. 30 (0.60 mm) -- a
No. 50 (0.30 mm) -- b
No. 60 (0.25 mm) -- a
No. 100 (0.15 mm) -- a Figure 3.3. Photograph of RCP-LS-IL material
No. 200 (0.075 mm) 8 4 used in laboratory testing.
a
sieve not specified
b
not tested for CA 6
of the expressway was built during the early to mid-1960s. The shown in Figure 3.2, approximates the typical dense-graded
constitutive aggregate may have been dolomite (a double car- base layer (DGBL) gradation and generally meets the Illinois
bonate of calcium and magnesium [CaMg(CO3)2]) and not DOT CA 6 gradation requirements but on the fine side (in-
limestone (which is a single carbonate of calcium [CaCO3]) be- dicated by the ± limits on the target gradation). Figure 3.3
cause most of the quarries produced dolomite during the con- shows the RCP-LS-IL material.
struction period. Calcium and magnesium have very similar
properties; the RCP from Illinois is referred to RCP-LS-IL,
RCP with Gravel
although the constitutive mineral could have been limestone
or dolomite. RCP with gravel (RCP-GV-LA) was obtained from the
The in-place PCC was processed on site to produce “CA 6” Louisiana DOT&D’s widening and rehabilitation project of
aggregate base course meeting the Illinois DOT gradation State Route 67/US 167 near Ruston. An on-site crushing plant
shown in Table 3.5. The as-received gradation of RCP-LS-IL, was used to produce RCP; PCC originated from rehabilitation
100
90
80
70
Percent Passing
60
50
40
30
20
RCP-LS-IL
10
Illinois DOT CA 6
0
10 1 0.1 0.01 0.001
Sieve Size Opening (inch)
Figure 3.2. RCP-LS-IL as-received gradation and typical Illinois DOT CA 6 gradation.
16
Table 3.6. Louisiana DOT&D gradation for

Class I and II coarse aggregate.

1.5 in (37.5 mm) 100
1.0 in (25 mm) 95 5
3/4 in (19 mm) 85 15
No. 4 (4.75 mm) 50 15
No. 40 (0.425 mm) 22 10
No. 200 (0.075 mm) 9 4
of I-20 near Monroe, Louisiana. The in-place PCC pavement

slabs were in fair to poor condition prior to transportation
to the crushing plant. Louisiana DOT&D allows the use of
100-percent RCP or in combination with an approved virgin Figure 3.5. Photograph of RCP-GV-LA material
aggregate. used in laboratory testing.
The existing PCC was crushed to produce RCP meeting the
gradation requirements shown in Table 3.6. The as-received
shipped to a central crushing plant in Sumter, South Carolina,
RCP-GV-LA gradation closely matches Louisiana DOT&D
for RCP production. PCC was crushed to meet the South Car-
and typical DGBL gradation requirements (see Figure 3.4).
olina DOT gradation for RCP base course shown in Table 3.7.
Figure 3.5 shows the RCP-GV-LA material.
The as-received gradation of RCP-GR-SC is shown in Fig-
ure 3.6. Figure 3.7 shows photograph of RCP-GR-SC material
RCP with Granite used during laboratory testing.
RCP with granite (RCP-GR-SC) was obtained from rehabil-
itation and widening of the aircraft parking apron at Colum-
RAP with Limestone
bia Airport in South Carolina. The in-place PCC slabs were in
fair condition and were removed as part of the apron expan- RAP with limestone (RAP-LS-MS) was obtained from a re-
sion and rehabilitation project. The removed PCC slabs were habilitation project on I-59 near Quitman, Mississippi. The
100
90
80
70
Percent Passing
60
50
40
30
20
RCP-GV-LA
10
Louisiana DOT&D Class I and II
0
10 1 0.1 0.01 0.001
Figure 3.4. RCP-GV-LA as-received and typical Louisiana DOT&D

gradations.
17
Table 3.7. Gradation requirements for

South Carolina DOT recycled PCC
base course.

2.0 in (50.8 mm) 100
1.5 in (37.5 mm) 95 - 100
1.0 in (25 mm) 70 - 100
1/2 in (12.5 mm) 48 - 75
No. 4 (4.75 mm) 30 - 50
No. 30 (0.60 mm) 11 - 30
No. 200 (0.075 mm) 0 - 12
in-place HMA surface was in good condition before milling;

the as-received RAP-LS-MS gradation is shown in Figure 3.8. Figure 3.7. Photograph of RCP-GR-SC material
Figure 3.9 shows the RAP-LS-MS material used in laboratory used in laboratory testing.
testing.
RAP with Granite RAP with Gravel

RAP with granite (RAP-GR-CO) was obtained by milling RAP with gravel (RAP-GV-LA) was obtained from Rayville,
2 inches (50 mm) from the surface of a deteriorated major Louisiana; the as-received gradation is shown in Figure 3.12.
arterial street in eastern Denver. The as-received gradation RAP-GV-LA had a slightly finer gradation than a typical
of the RAP-GR-CO material is shown in Figure 3.10. Figure OGDL gradation. RAP was obtained by milling, and the pave-
3.11 shows RAP-GR-CO material used during laboratory ment surface was in fair condition. Figure 3.13 shows RAP-
testing. GV-LA material used during laboratory testing.
100
90
80
70
Percent Passing
60
50
40
30
20
RCP-GR-SC
10 South Carolina DOT RCA Gradation
0
10 1 0.1 0.01 0.001
Figure 3.6. RCP-GR-SC as-received gradation and typical SC DGBL gradation.

100
90
80
70
Percent Passing
60
50
40
30
20
RAP-LS-MS
10
Mississippi DOT Crushed Stone Aggregate
0
10 1 0.1 0.01 0.00
Figure 3.8. RAP-LS-MS as-received and Mississippi DOT

aggregate gradations.
Figure 3.9. Photograph of RAP-LS-MS material

used for laboratory testing.
100
Typical NSA DGBL gradation

90
RAP-GR-CO
80
70
Percent Passing
60
50
40
30
20
10
0
10 1 0.1 0.01 0.001
Figure 3.10. RAP-GR-CO as-received gradation and typical NSA DGBL

gradation.
Figure 3.11. Photograph of RAP-GR-CO material
used for laboratory testing.
100
Typical NSA OGDL gradation
90
RAP-GV-LA
80
70
60
50
Percent Passing
40
30
20
10
0
10 1 0.1 0.01 0.001
Figure 3.12. RAP-GV-LA as-received and typical NSA OGDL gradation.
Figure 3.13. Photograph of RAP-GV-LA material used

for laboratory testing.
20
CHAPTER 4
Laboratory Test Program and Test Results
Laboratory Investigation terial are listed in Table 4.2. As indicated, the OMCs of two ma-
terials (50/50 blend of RCP-GR-SC DGBL#2 and RAP-GR-CO
RAP and RCP materials are generally used as an unbound 50/50 OGDL re-blend) were changed from the laboratory-
structural layer. Thus, most of the laboratory tests were con- determined values because of the free moisture observed dur-
ducted on samples meeting a target gradation similar to a ing repeated load triaxial testing.
DGBL; a few tests were conducted on samples prepared to a
gradation similar to an OGDL. The target gradations, shown
in Table 4.1, are based on typical gradations for virgin mate- Static Triaxial Test
rials as provided in the Aggregate Handbook (2), which were The static triaxial test was conducted in accordance with
adjusted based on results of the literature search and consid- AASHTO T 234 on each sample at confining stresses of 0, 5, and
eration of current practices. 15 psi (0, 34.5, 103.4 kPa) to determine the shear strengths at
RAP and RCP materials meeting the target gradations were OMC. Samples were prepared to approximately 95 percent of
procured to allow testing of recycled materials with gradations the maximum dry density values listed in Table 4.2. Table 4.3
similar to as-produced gradations. Table 4.2 shows the test shows the maximum deviator stress at various confining
conducted and the material combinations that were evaluated. stresses. Coarse DGBL gradation of virgin aggregate (DGBL#1)
had a higher maximum deviator stress compared to the finer
Results of Laboratory Tests DGBL gradation (DGBL#2); virgin OGDL had a lower maxi-
mum deviator stress compared to the DGBL gradations.
Grain Size Analysis Figure 4.6 shows the maximum deviator stress at 15 psi
All samples met the DGBL and OGDL gradation require- (103.4 kPa) confining pressure in ascending order. Overall,
ments. The gradations with corresponding target for as- RCP samples had the greatest maximum deviator stress,
received RCP DGBL, RAP OGDL, and RAP DGBL are shown followed by virgin aggregate materials and RAP samples.
in Figures 4.1, 4.2, and 4.3, respectively. The gradations for the Materials with as-received DGBL gradations also had a higher
blended virgin aggregate DGBL and virgin aggregate OGDL deviator stress compared to material with OGDL as-received
samples are shown in Figures 4.4 and 4.5, respectively. gradations. RAP and RCP with granite aggregate had a higher
RCP-GR-SC and RAP-GR-CO met the DGBL requirements maximum deviator stress, followed by materials with gravel
as-received. These materials were blended to OGDL gradation. and limestone.
The constitutive aggregate in material referred to as RCP-LS-
IL (in Figure 4.1 and subsequent figures) could have been
Repeated Load Triaxial Test Results
dolomite or limestone. Figure 4.4 shows two virgin aggregate
samples blended to meet DGBL requirements. The repeated load triaxial tests were conducted to obtain
a relative measure of the resistance of tested materials to
permanent deformation. The test procedure, described in
Moisture/Density Relations
detail in Appendix B (available as NCHRP Web-Only Docu-
Test specimens were prepared by compacting the RAP, RCP, ment 119 available on the webpage), is briefly discussed in
and blends in accordance with test method D of AASHTO Chapter 3. At each load level, 1,000 cycles were applied; the
T 180. The OMC and the maximum dry densities for each ma- deviator stress for the first two load levels was 10 and 20 psi
21
Table 4.1. Percent passing for laboratory testing.
Sieve size DGBL OGDL

1.50 inch (37.5 mm) 95 - 100 100
3/4 inch (19.0 mm) 70 - 89 70 - 95
3/8 inch (9.5 mm) 50 - 70 35 - 65
No. 4 (4.75 mm) 35 - 55 20 - 40
a
No. 16 (1.18mm) -- 0 - 10
No. 30 (0.6 mm) 12 - 25 -- a
a
No. 50 (0.3 mm) -- 0-5
a
No. 100 (0.15 mm) -- 0-3
No. 200 (0.075 mm) 0-7 0-1
a
Not recorded
(68.9 and 137.9 kPa) and increased by 20 psi (137.9 kPa) there- repeated load triaxial tests conducted on RCP-GR-SC in the
after until failure (defined by a permanent axial strain of dry condition (unsaturated) are shown in Figures 4.7 and 4.8.
10 percent) occurred or the load-frame limit was reached. Figure 4.7 shows the axial strain percent versus deviator
Tests were conducted on triplicate OGDL samples prepared at stress relationships for the first and last load increments at the
OMC and on triplicate DGBL samples prepared at OMC in the beginning and following the 1000th cycle of a repeated stress
saturated (wet) and unsaturated (dry) conditions. Results for loading increment. Figure 4.8 shows the magnitude of axial
Table 4.2. OMC and density data.
Maximum Dry OMC

Material Tested Tested Gradation
Density (pcf) (percent)
Virgin DGBL Blend#1 DGBL#1 150.1 7.4
Virgin DGBL Blend#2 DGBL#2 141.1 6.3
Virgin OGDL Blend OGDL 132.7 8.8
RAP-LS-MS OGDL 124.1 6.3
RAP-GV-LA OGDL 123.5 5.4
RAP-GR-CO DGBL#2 125.8 10.3
RCP-LS-IL DGBL#1 123.0 11.0
RCP-GV-LA DGBL#1 121.7 9.0
RCP-GR-SC DGBL#2 124.2 9.5
50/50 RAP-LS-MS OGDL 128.7 6.8
50/50 RAP-GV-LA OGDL 130.7 5.9
50/50 RAP-GR-CO DGBL#2 130.3 4.0
50/50 RCP-LS-IL DGBL#1 130.5 8.1
50/50 RCP-GV-LA DGBL#1 132.0 7.6
50/50 RCP-GR-SC DGBL#2 128.8 9.0 a
RAP-GR-CO 100%OGDL re-blend OGDL 123.7 5.6
RAP-GR-CO 50/50 OGDL re-blend OGDL 127.5 3.5
RCP-GR-SC 100%OGDL re-blend OGDL 120.2 9.0 b
RCP-GR-SC 50/50 OGDL re-blend OGDL 124.8 9.0
Notes: 1 pcf (pound/ft3) = 16.02 kg/m3
a
OMC was lowered from 13.0 percent to 9.0 percent because of the free moisture
observed during repeated load triaxial tests.
b
OMC was lowered from 15.4 percent to 9.0 percent because of the free moisture
observed during repeated load triaxial tests.
22
100
DGBL-Target
90
RCP-GV-LA
80
RCP-LS-IL
70 Virgin DGBL blend#2
Percent Passing
60 RCP-GR-SC
50
40
30
20
10
0
10 1 0.1 0.01 0.001
Figure 4.1. Gradation results of RCP as-received DGBL samples.
strain percent versus number of load cycles. Load increments, Table 4.5 lists the stress ratios at which a particular per-
applied axial stress for each load cycle, and corresponding manent strain (1, 3, 7, or 10 percent) occurred for each trip-
stress ratios are shown in Table 4.4. The stress ratio is defined licate sample. A higher stress ratio at lower permanent
as the ratio of major principal stresses (i.e., the ratio between strain indicates a material with more resistance to perma-
vertical and horizontal stresses); it is equal to the ratio be- nent deformation.
tween the axial applied stress and chamber confining pres- Stress ratios (average of three tests) obtained from dry
sure (i.e., σ1 /σ3). (tests conducted at OMC) repeated load triaxial tests are
100
OGDL-Target
90
RAP-GV-LA
80
RAP-LS-MS
70
Percent Passing
60
50
40
30
20
10
0
10 1 0.1 0.01 0.001
Figure 4.2. Gradation results of RAP as-received OGDL samples.

23
100
90 DGBL-Target
80
RAP-GR-CO
70
Percent Passing 60
50
40
30
20
10
0
10 1 0.1 0.01 0.001
Figure 4.3. Gradation results of as-received RAP DGBL samples.
shown in Figure 4.9 for RAP and Figure 4.10 for RCP and vir- lowed by virgin aggregate DGBL#1 and OGDL gradations.
gin aggregate. The material resistant to permanent deforma- Figures 4.11 and 4.12 show the stress ratios (average of three
tion has the highest stress ratio at the lowest permanent tests) for wet and dry tests on RAP and virgin aggregate sam-
strain. In dry tests, 50-percent blends of RAP with virgin ag- ples (DGBL gradations) and on RCP (DGBL gradations).
gregate exhibited the highest permanent deformation resist- Virgin aggregate DGBL#1 exhibited the least permanent
ance of RAP materials. RCP-GV-LA exhibited the highest strain in dry repeated load triaxial tests, and virgin aggregate
overall permanent deformation resistance in the dry test, fol- DGBL#2 the least permanent strain in wet tests. Of tested
100
90 DGBL-Target
Virgin-DGBL
80 Virgin DGBL blend#2
70
Percent Passing
60
50
40
30
20
10
0
10 1 0.1 0.01 0.001
Figure 4.4. Gradation results of virgin aggregate DGBL-blended sample.

24
100
90 OGDL-Target
80
Virgin-OGDL
70
Percent Passing
60
50
40
30
20
10
0
10 1 0.1 0.01 0.001
Figure 4.5. Gradation results of virgin aggregate OGDL-blended sample.
Table 4.3. Failure deviator stress.
Confining Pressure (psi)

Sample Identification Blend Φ c (psi) 0 5 15
Max. Deviator Stress, σd (psi)
Virgin DGBL Blend#1 DGBL#1 48.0 5.21 42.09 a 69.54 b 121.14
Virgin DGBL Blend#2 DGBL#2 49.0 2.08 11.20 44.21 106.40
Virgin OGDL Blend OGDL 45.0 2.78 27.17 a 50.59 b 83.51
RAP-LS-MS OGDL 38.5 4.17 26.86 a 41.10 b 65.30
RAP-GV-LA OGDL 39.0 2.78 23.56 a 37.82 b 61.73
RAP-GR-CO DGBL#2 41.0 2.08 11.20 32.20 63.96
RCP-LS-IL DGBL#1 46.0 5.56 39.16 a 66.36 b 103.20
RCP-GV-LA DGBL#1 48.0 1.39 25.39 a 55.06 b 96.05
RCP-GR-SC DGBL#2 52.0 2.78 19.26 56.33 129.58
50/50 RAP-LS-MS c OGDL -- -- -- -- --
50/50 RAP-GV-LA c OGDL -- -- -- -- --
50/50 RAP-GR-CO DGBL#2 42.0 1.74 6.77 33.52 69.74
50/50 RCP-LS-IL DGBL#1 50.0 2.08 12.45 50.65 106.22
50/50 RCP-GV-LA c DGBL#1 -- -- -- -- --
50/50 RCP-GR-SC DGBL#2 50.0 2.08 10.18 50.30 109.63
RAP-GR-CO 100% OGDL re-blend OGDL 52.0 1.74 8.96 31.49 128.94
RAP-GR-CO 50/50 OGDL re-blend OGDL 42.0 1.74 6.18 31.17 68.25
RCP-GR-SC 100% OGDL re-blend OGDL 50.0 2.78 16.88 52.24 111.24
RCP-GR-SC 50/50 OGDL re-blend OGDL 49.0 1.74 12.40 55.53 103.63
Notes: 1 psi = 6.9 kPa
a
Maximum deviator stress at confining stress of 3 psi
b
Maximum deviator stress at confining stress of 7 psi
c
Not tested
25
RCP-GR-SC
RAP-GR-CO R
Virgin (DGBL#1)
RCP-GR-SC R
RCP-GR-SC 50/50
Virgin (DGBL#2)
RCP-LS-IL 50/50
RCP-GR-SC 50/50 R
RCP-LS-IL
RCP-GV-LA
Virgin (OGDL)
RAP-GR-CO 50/50
RAP-GR-CO 50/50 R
OGDL
RAP-LS-MS
DGBL#2
RAP-GR-CO DGBL#1
RAP-GV-LA
0 20 40 60 80 100 120 140
Maximum Deviator Stress, psi (at 15 psi confining stress)
Figure 4.6. Maximum deviator stress in static triaxial tests.
180
160
140
Deviator Stress (psi)
120
100
80
60
40
Strain First Load Increment

20
Strain Last Load Increment
0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
9.5
10.0
10.5
11.0
Axial Strain (%)
Figure 4.7. Repeated load triaxial test results for RCP-GR-SC at OMC.
26
10
8
10% permanent strain reached at
7 cycle #677 of loading sequence #9
Axial Strain (%) 6
0
0
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
4,500
5,000
5,500
6,000
6,500
7,000
7,500
8,000
8,500
9,000
9,500
10,000
No. of Cycles (N)
Figure 4.8. Load repetitions versus permanent axial strain for RCP-GR-SC
at OMC.
materials, the 50-percent blend of RAP-GR-CO with virgin which the average permanent deformation was only 3.67 per-
aggregate was more resistant to permanent deformation in cent. Similar data are presented in Figure 4.14 for wet and dry
both the wet and dry repeated load triaxial tests compared to tests on DGBL gradations.
as-received RAP-GR-CO material. RCP-GV-LA exhibited
the highest permanent deformation resistance in the dry tests Resilient Modulus Test Results
of all RCP materials. In wet tests, the RCP-GR-SC material
showed better resistance to permanent deformation. Resilient modulus (MR) was obtained using repeated load
The number of load repetitions required to cause failure triaxial test data. MR values determined for different bulk
(10-percent permanent strain) was also used to evaluate re- stresses at an as-tested confining pressure of 15 psi (103.4 kPa)
sistance to permanent deformation (Figure 4.13). Virgin ag- are shown in Figure 4.15. The relationship is expressed by the
gregate DGBL exhibited the highest resistance to permanent equation:
deformation. The test was terminated after 10,000 cycles at
M R =K1θK2
where θ is bulk stress (psi)

Table 4.4. Stress ratios for repeated load triaxial test K1 and K2 are the experimental constant and coeffi-
load cycles. cient, respectively
Load Axial Stress
Load cycles Stress Ratio K1 and K2 values determined for the tested materials are
Increment psi kPa
1 0 - 1,000 10 68.9 0.67 listed in Table 4.6 in order of stiffness for the dry tests. The MR
2 1,001 - 2,000 20 137.9 2.00 values were calculated for 100 psi (689.5 kPa) bulk stress using
3 2,001 - 3,000 40 275.8 3.33 K1 and K2 values determined for dry and wet repeated load tri-
4 3,001 - 4,000 60 413.7 4.67
axial tests. Virgin aggregate has the greatest stiffness, followed
5 4,001 - 5,000 80 551.6 6.00
6 5,001 - 6,000 100 689.5 7.33
by the 50-percent blend of RAP-LS-MS, RCP-GV-LA, RAP-
7 6,001 - 7,000 120 827.4 8.67 LS-MS, and 50-percent blend of RAP-GR-CO. Most of the
8 7,001 - 8,000 140 965.3 10.00 materials tested in both and wet dry conditions did not show
9 8,001 - 9,000 160 1,103.2 11.33 significant reduction in stiffness in the wet condition. In
10 9,001 - 10,000 180 1,241.1 12.67
fact, the stiffness of RAP and RCP samples with granite in
27
Table 4.5. Stress ratios for different permanent strain levels.
Load increment for strain achieved and stress ratio (SR)

Material Sample Identification
1% σ1/σ3 3% σ1/σ3 7% σ1/σ3 10% σ1/σ3
Dry Test #1 5 6.0 8 10.0 4.8% strain at cycle #10,000
Virgin DGBL Dry Test #3 6 7.3 10 12.7 3.0% strain at cycle #10,000
Blend#1 Saturated Test #1 4 4.7 7 8.7 9 11.3 10 12.7
Saturated Test #2 4 4.7 7 8.7 10 12.7 10 12.7
Saturated Test #3 4 4.7 7 8.7 9 11.3 9 11.3
Dry Test #1 5 6.0 8 10.0 10 12.7 10 12.7
Dry Test #2 5 6.0 8 10.0 9 11.3 10 12.7
Virgin
Dry Test #3 5 6.0 8 10.0 9 11.3 10 12.7
DGBL
Blend#2 Saturated Test #1 5 6.0 7 8.7 8 10.0 8 10.0
Saturated Test #2 5 6.0 6 7.3 8 10.0 8 10.0
Saturated Test #3 5 6.0 7 8.7 8 10.0 8 10.0
Virgin Dry Test #1 6 7.3 8 10.0 9 11.3 9 11.3
OGDL Dry Test #2 5 6.0 8 10.0 9 11.3 9 11.3
Blend Dry Test #3 5 6.0 8 10.0 9 11.3 10 12.7
Dry Test #1 3 3.3 5 6.0 7 8.7 9 11.3
RAP-LS- Dry Test #2 3 3.3 5 6.0 7 8.7 9 11.3
MS Dry Test #3 3 3.3 5 6.0 7 8.7 8 10.0
100oF Test 3 3.3 5 6.0 7 8.7 9 11.3
Dry Test #1 3 3.3 5 6.0 7 8.7 8 10.0
RAP-GV-
LA Dry Test #2 3 3.3 5 6.0 7 8.7 8 10.0
Dry Test #3 4 4.7 5 6.0 7 8.7 8 10.0
Dry Test #1 3 3.3 5 6.0 8 10.0 10 12.7
Dry Test #2 3 3.3 5 6.0 8 10.0 10 12.7
Dry Test #3 3 3.3 5 6.0 8 10.0 9 11.3
RAP-GR-CO 100oF Test 2 2.0 6 7.3 8 10.0 10 12.7
Saturated Test #1 3 3.3 5 6.0 8 10.0 10 12.7
Saturated Test #2 3 3.3 5 6.0 8 10.0 9 11.3
Saturated Test #3 3 3.3 6 7.3 9 11.3 10 12.7
Dry Test #3 5 6.0 7 8.7 9 11.3 10 12.7
RCP-LS-IL
Saturated Test #1 5 6.0 8 10.0 9 11.3 10 12.7
Saturated Test #2 6 7.3 8 10.0 10 12.7 10 12.7
Saturated Test #3 6 7.3 8 10.0 9 11.3 10 12.7
Dry Test #1 8 10.0 2.8% strain, cycle #9,001, at > load cell capacity
Dry Test #3 6 7.3 8 10.0 10 12.7 10 12.7
RCP-GV-LA
Saturated Test #1 5 6.0 6 7.3 7 8.7 8 10.0
Saturated Test #2 5 6.0 6 7.3 8 10.0 8 10.0
Saturated Test #3 5 6.0 6 7.3 8 10.0 8 10.0
Dry Test #1 5 6.0 7 8.7 9 11.3 9 11.3
Dry Test #2 5 6.0 7 8.7 9 11.3 9 11.3
RCP-GR- Dry Test #3 5 6.0 7 8.7 9 11.3 10 12.7
SC Saturated Test #1 5 6.0 7 8.7 8 10.0 9 11.3
Saturated Test #2 5 6.0 6 7.3 8 10.0 9 11.3
Saturated Test #3 5 6.0 7 8.7 8 10.0 9 11.3
Dry Test #1 4 4.7 6 7.3 8 10.0 9 11.3
50/50 RAP- Dry Test #2 4 4.7 6 7.3 8 10.0 9 11.3
LS-MS Dry Test #3 4 4.7 6 7.3 8 10.0 9 11.3
100oF Test 4 4.7 6 7.3 8 10.0 9 11.3
Dry Test #1 4 4.7 6 7.3 8 10.0 8 10.0
50/50 RAP-
GV-LA Dry Test #2 4 4.7 6 7.3 8 10.0 9 11.3
Dry Test #3 4 4.7 6 7.3 8 10.0 8 10.0
(continued on next page)
28
Table 4.5. (Continued).
Load increment for strain achieved and stress ratio (SR)

Material Sample Identification
1% σ1/σ3 3% σ1/σ3 7% σ1/σ3 10% σ1/σ3
Dry Test #1 4 4.7 6 7.3 8 10.0 9 11.3
Dry Test #2 4 4.7 6 7.3 8 10.0 9 11.3
Dry Test #3 4 4.7 6 7.3 8 10.0 9 11.3
50/50 RAP-
100oF Test 3 3.3 5 6.0 8 10.0 9 11.3
GR-CO
Saturated Test #1 4 4.7 6 7.3 8 10.0 9 11.3
Saturated Test #2 4 4.7 6 7.3 8 10.0 9 11.3
Saturated Test #3 4 4.7 6 7.3 8 10.0 9 11.3
Dry Test #1 4 4.7 7 8.7 9 11.3 10 12.7
Dry Test #2 5 6.0 7 8.7 9 11.3 10 12.7
50/50 RCP- Dry Test #3 5 6.0 7 8.7 9 11.3 10 12.7
LS-IL Saturated Test #1 4 4.7 6 7.3 7 8.7 8 10.0
Saturated Test #2 5 6.0 6 7.3 8 10.0 9 11.3
Saturated Test #3 5 6.0 6 7.3 7 8.7 8 10.0
Dry Test #1 5 6.0 7 8.7 9 11.3 9 11.3
Dry Test #2 5 6.0 7 8.7 9 11.3 10 12.7
50/50 RCP- Dry Test #3 5 6.0 7 8.7 8 10.0 8 10.0
GV-LA Saturated Test #1 4 4.7 6 7.3 7 8.7 8 10.0
Saturated Test #2 5 6.0 7 8.7 8 10.0 9 11.3
Saturated Test #3 5 6.0 6 7.3 8 10.0 8 10.0
Dry Test #1 5 6.0 7 8.7 10 12.7 10 12.7
Dry Test #2 4 4.7 7 8.7 9 11.3 9 11.3
50/50 RCP- Dry Test #3 4 4.7 6 7.3 8 10.0 9 11.3
GR-SC Saturated Test #1 4 4.7 6 7.3 8 10.0 9 11.3
Saturated Test #2 4 4.7 6 7.3 8 10.0 8 10.0
Saturated Test #3 4 4.7 6 7.3 7 8.7 8 10.0
RAP-GR-CO Dry Test #1 3 3.3 5 6.0 7 8.7 9 11.3
100%OGDL Dry Test #2 3 3.3 5 6.0 8 10.0 9 11.3
re- blend Dry Test #3 3 3.3 5 6.0 7 8.7 9 11.3
RAP-GR-CO Dry Test #1 4 4.7 5 6.0 7 8.7 9 11.3
50/50 OGDL Dry Test #2 3 3.3 5 6.0 7 8.7 8 10.0
re-blend Dry Test #3 4 4.7 5 6.0 8 10.0 9 11.3
RCP-GR-SC Dry Test #1 4 4.7 6 7.3 8 10.0 8 10.0
100%OGDL Dry Test #2 5 6.0 7 8.7 8 10.0 9 11.3
re-blend Dry Test #3 5 6.0 7 8.7 8 10.0 9 11.3
RCP-GR-SC Dry Test #1 5 6.0 7 8.7 8 10.0 9 11.3
50/50 OGDL Dry Test #2 5 6.0 7 8.7 8 10.0 9 11.3
re-blend Dry Test #3 5 6.0 7 8.7 8 10.0 9 11.3
as-received condition increased slightly under wet condi- an aggregate’s degradation potential. Test results are listed in
tions. However, the stiffness of the 50-percent blend of RCP- Table 4.7; data reported for 50-percent blends were calculated
GV-LA, virgin DGBL#1, and RCP-GV-LA was reduced by using data from tests on as-received gradations.
15, 20, and 33 percent when tested in wet conditions,
respectively, suggesting that these materials may be suscepti-
Durability
ble to wet conditions.
Aggregate durability, when subjected to freeze-thaw cycles
in the presence of moisture, was determined using the Cana-
dian Freeze-Thaw test (MTO LS-614). Test results are listed in
Aggregate toughness was determined using the Micro-Deval Table 4.7; test results for the 50-percent blends were calculated
test (AASHTO TP 58-00), which provides an indication of using data from tests on as-received gradations.
29
Virgin OGDL Blend
Virgin DGBL Blend#1
Virgin DGBL Blend#2
50/50 RAP-GR-CO
Strain
50/50 RAP-GV-LA
10 percent
7 percent
50/50 RAP-LS-MS
3 percent
1 percent
RAP-GR-CO 50/50 OGDL reblend
RAP-GV-LA
RAP-GR-CO 100%OGDL reblend
RAP-GR-CO
RAP-LS-MS
0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00

Stress Ratio
Figure 4.9. Stress ratios for dry triaxial tests on RAP samples.
RCP-GV-LA
Virgin OGDL Blend
Virgin DGBL Blend#1
RCP-GR-SC 50/50 OGDL reblend
50/50 RCP-GV-LA
RCP-GR-SC
RCP-LS-IL
Strain
10 percent
Virgin DGBL Blend#2
7 percent
3 percent
RCP-GR-SC 100%OGDL reblend
1 percent
50/50 RCP-LS-IL
50/50 RCP-GR-SC
0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00

Stress Ratio
Figure 4.10. Stress ratios for dry triaxial tests on RCP and virgin samples.
30
Virgin DGBL Blend#1
Virgin DGBL Blend#2
Strain
10 percent - wet
7 percent - wet
3 percent - wet
50/50 RAP-GR-CO 1 percent - wet
10 percent - dry
7 percent - dry
3 percent - dry
1 percent - dry
RAP-GR-CO
0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00

Stress Ratio
Figure 4.11. Stress ratios for dry and wet triaxial tests on RAP and virgin samples.
50/50 RCP-GR-SC
50/50 RCP-LS-IL
RCP-LS-IL
RCP-GR-SC
50/50 RCP-GV-LA
Strain
RCP-GV-LA 10 percent - wet
7 percent - wet
3 percent - wet
1 percent - wet
10 percent - dry
7 percent - dry
Virgin DGBL Blend#1 3 percent - dry
1 percent - dry
Virgin DGBL Blend#2
0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00

Stress Ratio
Figure 4.12. Stress ratios for dry and wet triaxial tests on RCP samples.
31
RAP-GV-LA
50/50 RAP-GV-LA
RAP-LS-MS

50/50 RAP-GR-CO
50/50 RAP-LS-MS
50/50 RCP-GV-LA
RCP-GR-SC
Virgin OGDL Blend
50/50 RCP-GR-SC
RAP-GR-CO
50/50 RCP-LS-IL
Virgin DGBL Blend#2
RCP-LS-IL
RCP-GV-LA
Virgin DGBL Blend#1
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
Number of load repetitions for 10 percent permanent strain
Figure 4.13. Number of load repetitions at 10-percent permanent strain

(dry tests).
RCP-GV-LA
Virgin DGBL Blend#2
50/50 RCP-GV-LA
50/50 RCP-GR-SC
50/50 RCP-LS-IL
50/50 RAP-GR-CO
RCP-GR-SC
RAP-GR-CO
Virgin DGBL Blend#1
RCP-LS-IL
Wet Test 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
Dry Test Number of load repetitions for 10 percent permanent strain
Figure 4.14. Failure load repetitions for wet and dry tests.
32
100,000
Resilient Modulus, psi
10,000
10 100 1000
Bulk Stress, psi
Figure 4.15. Resilient modulus test results for RCP-GR-SC.
Table 4.6. Resilient modulus data from repeated load triaxial testing.
Dry Tests Wet Tests MR at θ=100 psi

Sample Identification
K1 K2 K1 K2 Dry Wet
Virgin DGBL Blend#1 6,831 0.4897 12,487 0.3113 65,153 52,361
Virgin OGDL Blend 18,172 0.2657 -- a -- a 61,785 -- a
Virgin DGBL Blend#2 13,348 0.3269 15,338 0.2876 60,139 57,672
50/50 RAP-LS-MS 10,567 0.3691 -- a -- a 57,836 -- a
RCP-GV-LA 9,717 0.3870 36,947 0.0112 57,749 38,900
RAP-LS-MS 8,611 0.4008 -- a -- a 54,545 -- a
50/50 RAP-GR-CO 14,559 0.2760 15,010 0.2656 51,906 51,007
50/50 RAP-GV-LA 14,897 0.2666 -- a -- a 50,856 -- a
RAP-GR-CO 100%OGDL re-blend 9,786 0.3560 -- a -- a 50,418 -- a
50/50 RCP-GV-LA 19,308 0.2074 23,343 0.1318 50,184 42,824
RAP-GR-CO 50/50 OGDL re-blend 9,337 0.3630 -- a -- a 49,692 -- a
RCP-LS-IL 14,243 0.2713 25,452 0.1389 49,691 48,245
RAP-GR-CO 6,437 0.4411 8,459 0.3876 49,078 50,420
50/50 RCP-LS-IL 15,791 0.2432 24,468 0.1318 48,397 44,892
RAP-GV-LA 15,003 0.2534 -- a -- a 48,202 -- a
50/50 RCP-GR-SC 9,980 0.3263 15,576 0.2205 44,852 43,004
RCP-GR-SC 50/50 OGDL re-blend 10,172 0.3084 -- a -- a 42,097 -- a
RCP-GR-SC 100%OGDL re-blend 21,591 0.1229 -- a -- a 38,024 -- a
RCP-GR-SC 16,085 0.1848 21,774 0.1258 37,676 38,860
Notes: a wet tests on these materials were not included in laboratory investigation.
1 psi = 6.895 kPa
33
Table 4.7. Percent loss for Micro-Deval and Canadian

Freeze-Thaw tests.
Percent Loss
Material Identification Canadian Freeze-
Micro-Deval
Thaw
50/50 RCP-LS-IL * 6.40 10.70
Virgin DGBL Blend#1 6.60 0.60
RAP-GV-LA 7.50 1.90
50/50 RCP-GV-LA * 7.80 1.40
Virgin OGDL Blend 8.10 0.90
50/50 RCP-GR-SC * 8.65 12.95
RCP-GR-SC 10.70 25.30
50/50 RAP-GR-CO * 12.65 5.05
50/50 RCP-GV-LA * 13.05 12.25
50/50 RAP-LS-MS * 13.20 0.80
RAP-LS-MS 18.30 0.70
RAP-GR-CO 18.70 9.50
RCP-LS-IL 19.40 22.00
RCP-GV-LA 19.50 23.90
Note: * Percent loss for 50-percent blends are averages of data for as-received gradations
Frost Susceptibility Plots of the surface dielectric constant value versus time for
the tested materials are shown in Appendix C. Figures 4.18 and
Tube suction tests were conducted to characterize the mois-
4.19 show the data for RCP-GV-LA and RAP-GV-LA, respec-
ture susceptibility properties of RAP, RCP, and virgin aggre-
tively; RCP-GV-LA had rapid water absorption compared to
gate materials. The test measures the moisture affinity of a
RAP-GV-LA. Test results are listed in Table 4.8.
granular material by subjecting the test specimens to a 10-day
capillary soak in a water bath, as described in Texas Test
Method 144 E (12).
Materials with a high affinity for water will imbibe signifi-
cant amounts of water through suction, sometimes resulting in
moisture contents higher than optimum after the capillary
soaking period and substantial amounts of unbound, or “free,”
water in the aggregate matrix. This unbound water will influ-
ence the material’s ability to resist both traffic loading and
freeze-thaw cycling. In this test, specimens are molded at
OMC; RAP samples were more difficult to mold as compared
to RCP samples. Figure 4.16 shows some of the molded samples.
To monitor the amount of free water, the tube suction test
measures the surface dielectric constant of the material,
which is an indication of the free water in the aggregate sys-
tem, and studies (12) have shown that materials with a sur-
RAP-GV-LA RAP-LS-MS RCP-LS-IL RCP-GV-LA
face dielectric constant value of greater than 10 after the
capillary soak can, in some environments, exhibit poor per-
formance in the field. Figure 4.17 shows the apparatus used
to make the surface dielectric measurements. Figure 4.16. Molded RAP and RCP samples.
34
22.0
Marginal Unacceptable
20.0
Surface Dielectric Constant Value

18.0
16.0
14.0
12.0
10.0
8.0
Acceptable
6.0
4.0
2.0
0.0
0.0 50.0 100.0 150.0 200.0 250.0
Time (hours)
Figure 4.19. Dielectric constant value curve with time

for RAP-GV-LA.
Table 4.8. Dielectric constant values from the tube

suction test.
Figure 4.17. Apparatus used to make Sample Identification Dielectric Constant Value
surface dielectric measurements. RCP-GV-LA 14.3
Virgin DGBL Blend#1 9.6
50/50 RCP-GV-LA 13.5
50/50 RCP-LS-IL 21.6
22.0
Marginal Unacceptable
50/50 RCP-GR-SC 10.0

20.0
Virgin DGBL Blend#2 10.6
Surface Dielectric Constant Value
18.0
RCP-LS-IL 16.3
16.0 50/50 RAP-GR-CO 3.7
14.0 RAP-GR-CO 3.3
12.0 RCP-GR-SC 12.1
10.0 Virgin OGDL Blend 8.0
8.0 50/50 RAP-LS-MS 3.2
Acceptable
6.0 RAP-LS-MS 2.1

4.0 50/50 RAP-GV-LA 3.6
2.0 RAP-GR-CO 100%OGDL re-blend 3.4
0.0 RAP-GR-CO 50/50 OGDL re-blend 3.9
0.0 50.0 100.0 150.0 200.0 250.0 RAP-GV-LA 2.0
Time (hours) RCP-GR-SC 50/50 OGDL re-blend 9.5
RCP-GR-SC 100%OGDL re-blend 10.4
Figure 4.18. Dielectric constant value curve with time
for RCP-GV-LA. Note: Acceptance criterion is dielectric constant value of 10 or less.
35
CHAPTER 5
Analysis of Test Data
Selection of Performance-Based loss for different recycled materials, thus indicating that the
Test Methods test differentiates between different materials.
Laboratory test data were analyzed to identify the tests ap-

propriate for identifying RAP and RCP materials intended Durability
for use as unbound pavement layers, singularly or in combi- Recycled materials and virgin aggregate durability charac-
nation with other materials. teristics were determined using the Canadian Freeze-Thaw
An adequate test method is expected to measure parame- test (MTO LS-614); percent material loss for different aggre-
ters that influence performance and should be capable of dif- gates are shown in Figure 5.2. These data show that RCP sam-
ferentiating between sources, types, and blends of recycled ples and 50-percent blends of RCP with virgin aggregate had
aggregate. the largest material loss. It appears that the results are affected
by the production of excess fines from recycled materials dur-
ing testing, resulting from the disintegration of the cement
Toughness and Abrasion Resistance paste on the aggregate particles. Test results show the differ-
Recycled materials and virgin aggregate toughness and ence between different recycled materials and their blends
abrasion resistance characteristics were determined using the with virgin aggregate. Results of statistical evaluation of test
Micro-Deval test (AASHTO TP 58-00); percent loss for the data are shown in Table 5.2.
different materials are shown in Figure 5.1. RCP and RAP in
as-received condition exhibited more material loss than vir- Frost Susceptibility
gin aggregate material or 50-percent blends of recycled ma-
Frost susceptibility of the recycled and virgin aggregates was
terials with virgin aggregate. For the highest material loss, it
determined using the tube suction test. Aggregates were con-
appears that the test results were affected by the amount of
sidered acceptable, marginal, or unacceptable if the dielectric
fines produced during testing. The excess fines were not caused
constant was less than 10, between 10 and 16, and greater than
by aggregate degradation; therefore, test results are not appro-
16, respectively (12).
priate. Aggregates that exhibit material loss of 17 percent or
The order of recycled and virgin aggregate materials with
more have shown fair or poor field performance. (1, 13)
respect to dielectric constant value is shown in Table 5.3. All
The following two statistical hypotheses were used to
RAP samples met the dielectric constant criterion; some RCP
determine whether the test method differentiated between
samples failed this criterion.
RAP, RCP, and virgin aggregates at a statistically significant
The statistical test results are shown in Table 5.4. The tube
level (5 percent): suction test produced different results for the different material
types.
Null hypothesis, HO: MeanRAP = MeanVirgin
Alternate hypothesis, HA: MeanRAP 苷 MeanVirgin
Static Triaxial Test
Hypothesis testing results on Micro-Deval test data are Figure 5.3 shows the average maximum deviator stress at
shown in Table 5.1. These results indicate different material 15 psi (103.4 kPa) confining pressure sorted by aggregate type,
36
RCP-GV-LA
RCP-LS-IL
RAP-GR-CO
RAP-LS-MS
50/50 RAP-LS-MS
50/50 RCP-GV-LA
50/50 RAP-GR-CO
RCP-GR-SC
50/50 RCP-GR-SC
Virgin OGDL Blend
50/50 RCP-GV-LA
RAP-GV-LA
Virgin DGBL Blend#1
50/50 RCP-LS-IL
0 2 4 6 8 10 12 14 16 18 20
Percent Material Loss
Figure 5.1. Results of Micro-Deval tests.
tested gradation, and material tested. The test yielded differ- RCP, RCP-GV, and RCP-LS materials exhibited lower fail-
ent results for different conditions, although similar results ure strengths than those for the virgin aggregate material or
were obtained for materials containing limestone and gravel. the 50-percent blend with virgin aggregate. However, the fail-
Statistical analysis of test data also showed similar trends. ure strength of RCP material with granite increased with in-
Test results were different for RAP and RCP samples with a creased recycled material content.
p-value of 0.027. However, other comparisons indicated Results of the static triaxial test at 15 psi (103.4 kPa) con-
mixed results, probably due to limited test data. fining pressure indicated differences between RAP and RCP
Figures 5.4 and 5.5 show the maximum deviator stress at and between recycled materials containing different aggre-
15 psi (103.4 kPa) confining pressure for different percentages gate types.
of RAP and RCP recycled materials, respectively. For RAP
samples, the results of the static triaxial strength on as-received
Repeated Load Triaxial Test
materials indicated a decrease in deviator stress with increasing
recycled material content but not for the re-blended OGDL Figure 5.6 shows the “mean” maximum deviator stress at
gradation. 15 psi (103.4 kPa) confining pressure sorted by aggregate
Table 5.1. Statistical assessment of Micro-Deval test data at

5 percent test significance.
Null Hypothesis, Ho p - Value Remarks

0.030 Test method differentiated between RAP and
MeanRAP = MeanVirgin
(< 5%) virgin aggregate
0.060 Test method did not differentiate between RCP
MeanRCP = MeanVirgin
(> 5%) and virgin aggregate.
0.006 Test method differentiated between recycled
MeanRecycled = MeanVirgin
(< 5%) and virgin aggregate
0.655 Test method did not differentiate between 100
MeanRAP100% = MeanRAP50%
(> 5%) percent and 50 percent RAP samples.
MeanRCP100% = MeanRCP50%
(> 5%) percent and 50 percent RCP samples.
0.537 Test method did not differentiate between RAP
MeanRAP = MeanRCP
(> 5%) and RCP samples.
MeanRAP50% = MeanRCP50%
(> 5%) percent RAP and 50 percent RCP samples.
(> 5%) percent RAP and 100 percent RCP samples.
37
RCP-GR-SC
RCP-GV-LA
RCP-LS-IL
50/50 RCP-GR-SC
50/50 RCP-GV-LA
50/50 RCP-LS-IL
RAP-GR-CO
50/50 RAP-GR-CO
RAP-GV-LA
50/50 RCP-GV-LA
Virgin OGDL Blend
5050 RAP-LS-MS
RAP-LS-MS
Virgin DGBL Blend#1
0 5 10 15 20 25 30
Figure 5.2. Canadian Freeze-Thaw test results.
type, tested gradation, and material for tests conducted on un- pared with RAP blended with virgin aggregate material (50-
saturated and saturated samples (OGDL samples were tested percent blend). Tests conducted on RAP-GR-CO material in
under unsaturated conditions only). Test results indicate dif- both saturated and unsaturated conditions had similar results.
ferences between different conditions, although the differences Figures 5.8 and 5.9 show the results of repeated load tri-
between saturated and unsaturated tests were relatively small. axial tests on RCP samples. In general, there was decrease in de-
Figure 5.7 shows the results of repeated load triaxial tests viator stress when RCP samples were tested in the saturated
conducted at 15 psi (103.4 kPa) confining pressure. The RAP condition. However, the 100-percent RCP samples had rela-
samples with granite exhibited relatively higher deviator stress tively higher deviator stress than the 50-percent blend of RCP
in as-received gradation compared to when blended with with virgin aggregate. For most samples tested in the unsatu-
50-percent virgin aggregate. The capacity of RAP samples rated condition, there was a decrease in deviator stress with
with limestone and granite aggregate to resist permanent an increase in recycled material content. When tested under
deformation decreased with an increase in the recycled ma- saturated conditions, 100-percent RCP samples exhibited
terial content. RAP with granite aggregate exhibited higher higher deviator stress relative to RCP samples composed of
deviator stress by itself (100-percent recycled material) com- 50-percent virgin aggregate. The mean deviator stress of the
Table 5.2. Statistics for Canadian Freeze-Thaw data at 5-percent

test significance.

0.165 Test method did not differentiate between
(> 5%) RAP and virgin aggregate.
0.004 Test method differentiated between RCP and
(< 5%) virgin aggregate.
0.003 Test method differentiated between recycled
MeanRecycled = MeanVirgin
(< 5%) and virgin aggregates.
MeanRAP100% = MeanRAP50%
(> 5%) 100-percent and 50-percent RAP samples.
0.009 Test method differentiated between 100-
MeanRCP100% = MeanRCP50%
(< 5%) percent and 50-percent RCP.
MeanRAP = MeanRCP
(< 5%) RCP samples.
Test method did not differentiate between
0.142
MeanRAP50% = MeanRCP50% 50-percent RAP and 50-percent RCP
(> 5%)
samples.
0.012 Test method differentiated between 100-
(< 5%) percent RAP and 100-percent RCP samples.
38
Table 5.3. Tube suction test results.
Material Tested Dielectric Constant Value Rating

RAP-GV-LA 2.0 Acceptable
RAP-LS-MS 2.1 Acceptable
50/50 RAP-LS-MS 3.2 Acceptable
RAP-GR-CO 3.3 Acceptable
RAP-GR-CO 100%OGDL re-blend 3.4 Acceptable
50/50 RAP-GV-LA 3.6 Acceptable
50/50 RAP-GR-CO 3.7 Acceptable
RAP-GR-CO 50/50 OGDL re-blend 3.9 Acceptable
Virgin OGDL Blend 8.0 Acceptable
RCP-GR-SC 50/50 OGDL re-blend 9.5 Acceptable
Virgin DGBL Blend#1 9.6 Acceptable
50/50 RCP-GR-SC 10.0 Acceptable
RCP-GR-SC 100%OGDL re-blend 10.4 Marginal
Virgin DGBL Blend#2 10.6 Marginal
RCP-GR-SC 12.1 Marginal
50/50 RCP-GV-LA 13.5 Marginal
RCP-GV-LA 14.3 Marginal
RCP-LS-IL 16.3 Unacceptable
50/50 RCP-LS-IL 21.6 Unacceptable
Table 5.4. Statistical test results on Tube-Suction test data

at 5-percent test significance.

0.009 Test method differentiated between RAP
(< 5%) and virgin aggregate.
0.033 The test method differentiated between RCP
0.000 The test method differentiated between RCP
MeanRAP = MeanRCP
(< 5%) and RAP samples.
120
107 109
104 104
98
Maximum Deviator Stress, psi
96
100 89
87 86
76
80
60
40
20
0
Virgin
Granite
Gravel
Limestone
DGBL#1
DGBL#2
OGDL
RCP
Virgin
RAP
Aggregate Type Gradation Tested Material
Figure 5.3. Static triaxial test results at 15 psi (103.4 kPa) confining pressure.
39
140
120
Maximum deviator stress, psi

100
80
60
40
RAP-GR-CO (DGBL#2)
RAP-GR-CO Reblend (OGDL)
20
RAP-GV-LA (OGDL)
RAP-LS-MS (OGDL)
0
0 50 100
Recycled material content, percent
Figure 5.4. Maximum deviator stress versus recycled material content (RAP).
unsaturated test appeared to be unaffected by the amount of hibited higher maximum deviator stress than RAP material.
RCP in the test sample (165 psi [1138 kPa]) for 50-percent RCP When tested in the saturated condition, the RCP and virgin
blends with virgin aggregate compared to 163 psi (1124 kPa) for aggregate showed higher maximum deviator stress than RAP
100-percent RCP. material.
Statistical significance test results, shown in Table 5.5, in- The materials selected for laboratory tests were expected to
dicated similar trends. The test method correctly differenti- provide a range of expected performance as indicated by
ated between different materials. shear strength. At 1-percent strain, 100-percent RAP material
Figures 5.10, 5.11, and 5.12 show the order of maximum had the lowest strength, followed by 50 percent blends of
deviator stress of recycled materials tested in the unsaturated RAP with virgin aggregate. The 100 percent RCP and virgin
condition in the repeated load triaxial strength at 1-, 2-, and aggregate samples had the highest strengths; the 50 percent
3-percent strain, respectively. Figure 5.13 shows the shear RCP blends with virgin aggregate had the second highest
strengths at 1-, 2-, and 3-percent strain for tests conducted in shear strengths. Shear strengths estimated at 3 percent strain
the saturated condition. Overall, RCP and virgin aggregate ex- provided somewhat different order.
140
Maximum deviator stress, psi
120
100
80 RCP-GR-SC (DGBL#2)
RCP-GR-SC Reblend (OGDL)
RCP-GV-LA (DGBL#1)
RCP-LS-IL (DGBL#1)
60
0 50 100
Figure 5.5. Maximum deviator stress versus recycled material content (RCP).
40
Unsaturated Tests
200
Saturated Tests
178
177
176
176
173
180
166
167
166
163
162
161
159
158
158
157
157
157
154
160
139
140
120
100
80
60
40
20
Gravel
DGBL#1
DGBL#2
OGDL
Granite
RCP
Virgin
Virgin
Limestone
RAP
Agg.
Agg.
Aggregate Type Gradation Tested Material
Figure 5.6. Repeated load triaxial test results at 15 psi confining pressure.
The slopes of the deviator stress versus axial strain curve Resilient Modulus Test
obtained during the load and unload cycles in the repeated
load triaxial testing could provide an indication of the per- The resilient modulus (or stiffness) was estimated at dif-
formance potential of recycled materials. Figure 5.14 shows ferent bulk stresses from data obtained during repeated load
typical load/unload curves for RCP and RAP materials. Dur- triaxial tests; results at the bulk stress of 100 psi (689.5 kPa)
ing static triaxial testing, these materials failed at about 4- to are shown in Figure 5.16. RCP-GR-SC was the least stiff ma-
5-percent strain. However, in repeated load triaxial testing, terial. Order-based saturated test results are shown in Figure
the sample sustained a higher load due to aggregate interlock 5.17. Generally, virgin aggregate and 50-percent blends of re-
and resistance characteristics. Good quality materials indi- cycled materials with virgin aggregate exhibited higher stiff-
cate a large slope (change in deviator stress per unit perma- ness than 100-percent recycled materials. Statistical analysis
nent strain) or low curvature at test initiation. The order of of resilient modulus data, shown in Table 5.6, indicate that
tested materials based on initial slope is shown in Figure 5.15. test data reveal differences between different materials.
190
Virgin aggregate
180 50/50 Blend
100% Recycled
170
160
150
140
130
120
110
100
RAP-GR-CO RAP-GR-CO RAP-GR-CO RAP-GV-LA RAP-LS-MS
(DGBL#2) (DGBL#2-Sat) (OGDL) (OGDL) (OGDL)
Figure 5.7. Repeated load triaxial test results for RAP samples.
41
190
180

170
160
150
140 RCP-GR-SC (DGBL)

RCP-IL-LS (DGBL)
RCP-SC-GR (OGDL)
130
RCP-GV-LA (DGBL-Sat)
120
0 50 100
Figure 5.8. Repeated load triaxial test results for RCP samples.
190
180
170
160
150
140
RCP-IL-LS (DGBL-Sat)
130 RCP-GR-SC (DGBL-Sat)
RCP-GV-LA (DGBL)
120
0 50 100
Figure 5.9. Repeated load triaxial test results for RCP-DGBL samples.
Table 5.5. Statistics for repeated load triaxial test data at 5-percent test
significance.
Test Condition Null Hypothesis, Ho p - Value Remarks

0.007 Test method differentiated between RCP
At OMC MeanRAP = MeanVirgin
MeanRAP = MeanRCP
(< 5%) and RCP samples.
(> 5%) RCP and virgin aggregate.
Saturated MeanRAP = MeanVirgin
(> 5%) RAP and virgin aggregate.
MeanRAP = MeanRCP
(< 5%) and RCP samples.
.019 Test method differentiated between 50
MeanRAP50% = MeanVirgin
OMC and (< 5%) percent RAP and virgin aggregate.
saturated 0.048 Test method differentiated between 50
MeanRCP50% = MeanVirgin
(< 5%) percent RCP and virgin aggregate.
42
RAP-LS-MS
RAP-CO-GR (OGDL)
RAP-CO-GR
RAP-GV-LA
RAP-CO-GR (OGDL) 50
RAP-LS-MS (50)
RAP-CO-GR (50)
RAP-GV-LA (50)
RCP-GR-SC (50)
RCP-GR-SC (OGDL) 50
RCP-GR-SC (OGDL)
RCP-LS-IL (50)
RCP-LS-IL
RCP-GR-SC
RCP-GV-LA (50)
Virgin #1
Virgin #2
Virgin OGDL
RCP-GV-LA
0 20 40 60 80 100 120 140

Maximum Deviator Stress at 1% strain, psi
Figure 5.10. Shear strength at 1-percent strain in repeated load triaxial test.
RAP-LS-MS
RAP-CO-GR (OGDL)
RAP-CO-GR
RAP-CO-GR (OGDL) 50
RAP-GV-LA
RAP-LS-MS (50)
RAP-CO-GR (50)
RAP-GV-LA (50)
RCP-GR-SC (OGDL)
RCP-GR-SC (OGDL) 50
RCP-GR-SC (50)
RCP-LS-IL (50)
RCP-GR-SC
RCP-GV-LA (50)
Virgin #2
RCP-LS-IL
Virgin OGDL
Virgin #1
RCP-GV-LA
0 20 40 60 80 100 120 140 160

43
RAP-LS-MS
RAP-CO-GR (OGDL)
RAP-CO-GR (OGDL) 50
RAP-CO-GR
RAP-GV-LA
RAP-LS-MS (50)
RAP-CO-GR (50)
RAP-GV-LA (50)
RCP-GR-SC (OGDL)
RCP-GR-SC (OGDL) 50
RCP-GR-SC (50)
RCP-LS-IL (50)
RCP-GR-SC
RCP-GV-LA (50)
Virgin #2
Virgin OGDL
RCP-LS-IL
RCP-GV-LA
Virgin #1
0 20 40 60 80 100 120 140 160 180

1% strain
RAP-CO-GR
2% strain
RAP-CO-GR (50) 3% strain
RCP-GR-SC (50)
RCP-LS-IL (50)
RCP-GV-LA (50)
RCP-GV-LA
Virgin #2
RCP-GR-SC
Virgin #1
RCP-LS-IL
0 20 40 60 80 100 120 140

Maximum Deviator Stress at 1, 2, 3 % strain, psi
Figure 5.13. Shear strength materials tested saturated in repeated load triaxial test.
44
180
160
140
Deviator Stress (psi)
120
100
80
60
40 RCP-GV-LA-Strain First Load Increment

RCP-GV-LA-Strain Last Load Increment
20 RAP-LS-MS-Strain First Load Increment
RAP-LS-MS-Strain Last Load Increment
0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0
Axial Strain (%)
Figure 5.14. Comparison of load/unload curves for typical RCP and

RAP materials.
RCP-LA-GV
Virgin OGDL
RCP-SC-GR-50-50 W
RCP-SC-GR-OGDL-50-50
RCP-GR-SC-OGDL
RCP-LA-GV W
RCP-IL-LS W
RCP-LA-GV-50-50
Virgin DGBL
RCP-IL-LS-50-50
RCP-SC-GR-OGDL-50-50
RCP-IL-LS-50-50
RCP-IL-LS
RAP-LA-GV-50-50
Virgin DGBL #2
RCP-LA-GV-50-51
Virgin DGBL #3
RCP-SC-GR-50-50
RCP-SC-GR-50-51
Virgin DGBL W
RAP-CO-GR-50-50
RAP-LA-GV
RAP-MS-LS-50-50
RAP-CO-GR-OGDL-50-50
RAP-CO-GR
RAP-CO-GR-50-50
RAP-CO-GR-OGDL
RAP-MS-LS
RAP-CO-GR W
0 20 40 60 80 100 120 140 160 180
Initial Slope (percent change in deviator stress per unit permanent strain)
Figure 5.15. Order of materials based on initial load-strain slope.

45
RCP-GR-SC
50/50 RCP-GR-SC OGDL reblend
50/50 RCP-GR-SC
RAP-GV-LA
50/50 RCP-LS-IL
RAP-GR-CO
RCP-LS-IL
50/50 RAP-GR-CO OGDL reblend
50/50 RCP-GV-LA
RAP-GR-CO OGDL reblend
50/50 RAP-GV-LA
50/50 RAP-GR-CO
RAP-LS-MS
RCP-GV-LA
50/50 RAP-LS-MS
Virgin DGBL Blend#2
Virgin DGBL Blend
Virgin DGBL Blend#1
0 10 20 30 40 50 60 70
Stiffness, Resilient Modulus (ksi)
Figure 5.16. Stiffness estimate using repeated load triaxial data at 100 psi
bulk stress.
Test Method Selection Summary characterize recycled materials. Based on results of the labo-
ratory investigation, the following tests were found to relate
The performance potential of an unbound pavement layer to performance:
depends on its dry and wet shear strength, resistance to
freeze-thaw (durability), toughness, and frost susceptibility.
These properties were evaluated using selected tests in a lab- • Screening tests for sieve analysis and the moisture-density
oratory investigation. Also, screening tests were conducted to relationship,
RCP-GR-SC
RCP-GV-LA
50/50 RCP-GV-LA
50/50 RCP-GR-SC
50/50 RCP-LS-IL
RCP-LS-IL
RAP-GR-CO
50/50 RAP-GR-CO
Virgin DGBL Blend#1
Virgin DGBL Blend#2
0 10 20 30 40 50 60 70
Stiffness, Resilient Modulus (ksi)
Figure 5.17. Estimated stiffness at 100 psi bulk stress (sat. repeated load
triaxial test).
46
Table 5.6. Statistics for Resilient Modulus data at 5 percent test significance.
Test Condition Null Hypothesis, Ho p - Value Remarks

MeanRCP = MeanRAP *
(> 5%) and RAP.
At OMC 0.000 Test method differentiated between RCP and
0.028 Test method differentiated between RAP and RCP
MeanRCP50 = MeanRAP50
(< 5%) samples containing 50-percent virgin aggregate.
(> 5%) percent RCP and 100-percent RAP.
MeanRCP = MeanRAP *
(> 5%) and RAP.
0.077 Test method did not differentiate between RAP
Saturated MeanRAP = MeanVirgin
(> 5%) and virgin aggregate.
0.020 Test method differentiated between RCP and
MeanRCP = MeanRAP *
(< 5%) RCP.
OMC and 0.000 Test method differentiated between RCP and
Saturated (< 5%) virgin aggregate.
0.009 Test method differentiated between RAP and RCP
(< 5%) samples containing 50-percent virgin aggregate.
(> 5%) percent RCP and 100-percent RAP.
* Includes 100-percent and 50-percent blend recycled material samples.
Table 5.7. Significance level of intended use on aggregate performance

potential.
Temperature Moisture Traffic

Condition Condition High Medium Low
High 4 4 3
Freezing
Low 4 3 2
High 3 2 2
Non Freezing
Low 3 2 1
Scale of 1 to 4 with 4 = Most significance, 1 = least significant
• The Micro-Deval test for toughness, • Low traffic (<100,000 ESALs/year),

• Resilient modulus for stiffness, • Medium traffic (100,000–1,000,000 ESALs/year), and
• Static triaxial and repeated load at OMC and saturated for • High traffic (>1,000,000 ESALs/year).
shear strength, and
The climatic conditions of moisture (high/low) and tem-
• Tube suction test for frost susceptibility.
perature (freezing/not freezing) are based on the AASHTO
definitions (14). Table 5.7 shows the significance levels of
Selection of Recycled Materials traffic, moisture, and climate combinations on a scale of 1 to
for Intended Use 4, where 4 is most significant and 1 least significant on aggre-
gate performance potential.
Recycled materials can be selected for use in a particular traf- Recycled materials could also be selected for use in a partic-
fic and climatic condition. Rangarajy et al. (13) developed an ular pavement structure (e.g., doweled PCC, undoweled PCC,
approach for evaluating aggregates using selected test param- or HMA at various traffic levels and climates) for a particular
eters, performance ratings, and traffic and climatic categories. base/subbase application (e.g., strength layer or construction
In this approach, tests are conducted in sequence and results or drainage layer). Different properties would be required of
are compared to suggested performance levels for specific traf- recycled materials for each unique situation. This level of detail
fic and climatic ranges. Three traffic levels are proposed: has not been considered in this research.
47
Table 5.8. Recommended tests and test parameters for levels of intended use.
Traffic H M H L M L
Tests
(Test Parameters)
Moisture H L H L H L H L
Temperature F NF F NF
Micro-Deval Test (percent loss) < 5 percent < 15 percent < 30 percent < 45 percent
Tube Suction Test (dielectric constant) 7 10 15 20
Static Triaxial Test OMC, c = 5psi 100 psi 60 psi 25 psi Not required
(Max. deviator stress)
Sat. c = 15psi 180 psi 135 psi 60 psi Not required
Repeated Load Test OMC, c = 15psi 180 psi 160 psi 90 psi Not required
(Failure deviator stress)
Sat. c = 15psi 180 psi 160 psi 60 psi Not required
Stiffness Test (Resilient modulus) 60 ksi 40 ksi 25 ksi Not required
Proposed ranges for selected test parameters that relate to shown Figure 5.18, indicate that recycled materials are generally
performance are shown in Table 5.8 for various levels of cli- appropriate for use in medium to low traffic conditions in non-
matic and traffic condition. These ranges determine the traf- freezing climates with low and high moisture contents. RCP-
fic and climatic conditions where these recycled materials GR and RAP-GV seem appropriate for use in high traffic areas
and their blends can be used. However, results from acceler- with non-freezing temperatures or in low and medium traffic
ated pavement tests and/or in-service test pavement evalua- areas in freezing climates with low moisture conditions.
tions are needed to confirm or refine these ranges. Adding virgin aggregate to recycled materials improves the
performance potential (based on the toughness test). For ex-
ample, virgin aggregates and 50-percent blend of recycled
Selection Based on Toughness Test
materials with virgin aggregate are appropriate for use in low
Recycled materials and virgin aggregate toughness and abra- and medium traffic areas in freezing climates with low mois-
sion resistance characteristics were evaluated using the Micro- ture conditions or high traffic areas with non-freezing tem-
Deval test. The test results and recommended test parameter, peratures with high or low moisture conditions. None of the
Significance Level 4 3 2 and less
RCP-GV-LA
RCP-LS-IL
RAP-GR-CO
RAP-LS-MS
50/50 RAP-LS-MS
50/50 RCP-GV-LA
50/50 RAP-GR-CO
RCP-GR-SC
50/50 RCP-GR-SC
Virgin OGDL Blend
50/50 RCP-GV-LA
RAP-GV-LA
Virgin DGBL Blend#1
50/50 RCP-LS-IL
0 2 4 6 8 10 12 14 16 18 20
Figure 5.18. Performance potential based on toughness (Micro-Deval) test.

48
tested materials (virgin, recycled or 50-percent blend of virgin Selection Based on Repeated Load
and recycled materials) are appropriate for use in perfor- Triaxial Test
mance significance level 4 (high traffic locations with freezing
temperatures and low and high moisture conditions). Failure deviator stress for repeated load triaxial tests
conducted at OMC, shown in Figure 5.21, indicate that vir-
Selection Based on Frost Susceptibility Test gin aggregates are appropriate for use in high traffic condi-
tions (significance level 4). RCP and 50-percent RCP blend
Frost susceptibility of different recycled and virgin aggre- with LS and GR are appropriate for use in conditions rep-
gates was determined using the tube suction test; results are resenting significance level 3 (i.e., high traffic level in non-
shown in Figure 5.19. The results indicate that RCP materials freezing temperatures, medium traffic level in freezing
are appropriate for use only in performance significance level 2 temperature in the presence of low moisture, and low traf-
(medium traffic no freezing) and level 1 (low traffic, no freez- fic level in freezing temperatures). RAP and 50-percent
ing, low moisture). Blending RCP with virgin aggregate in- RAP blends are generally appropriate for use in conditions
creased the performance potential to the next level, and thus representing significance level 2.
would be appropriate for use in high traffic (no freezing) and Failure deviator stress for repeated load triaxial tests
medium traffic (freezing with low moisture condition). RAP conducted in saturation condition, shown in Figure 5.22,
and 50-percent blends with virgin aggregate are appropriate indicate that RAP-GR and RCP-LS are appropriate for use
for use in high traffic conditions. in high moisture conditions with low or medium traffic
and non-freezing temperatures. The other materials are
Selection Based on Static Triaxial Test appropriate for use in conditions representing significance
The results of the static triaxial test, shown in Figure 5.20, level 2.
indicate that most of the RCP materials and their blends with
virgin aggregate are appropriate for use in extreme traffic and Selection Based on Material Stiffness
climatic conditions (significance level 4). RAP, on the other
hand, is appropriate for use in conditions representing sig- Most recycled materials and 50-percent blends with virgin
nificance level 3 (i.e., high traffic level in non-freezing tem- aggregate were shown to be appropriate for use in conditions
peratures, medium traffic level in freezing temperature in the representing significance level 3, as shown in Figure 5.23. Vir-
presence of low moisture, and low traffic level in freezing gin aggregates were shown to be appropriate for use in con-
temperatures). ditions representing significance level 4.
Significance Level 4 3 2 1
50/50 RCP-LS-IL
RCP-LS-IL
RCP-GV-LA
50/50 RCP-GV-LA
RCP-GR-SC
Virgin DGBL Blend#2
50/50 RCP-GR-SC
Virgin DGBL Blend#1
Virgin OGDL Blend
50/50 RAP-GR-CO
50/50 RAP-GV-LA
RAP-GR-CO
50/50 RAP-LS-MS
RAP-LS-MS
RAP-GV-LA
0 5 10 15 20 25
Final Dielectric Constant
Figure 5.19. Performance potential based on frost susceptibility.

49
Significance Level 1* 2 3 4
RCP-GR-SC
RAP-GR-CO R
Virgin (DGBL#1)
RCP-GR-SC R
RCP-GR-SC 50/50
Virgin (DGBL#2)
RCP-LS-IL 50/50
RCP-GR-SC 50/50 R
RCP-LS-IL
RCP-GV-LA
Virgin (OGDL)
RAP-GR-CO 50/50
RAP-GR-CO 50/50 R
RAP-LS-MS
RAP-GR-CO
RAP-GV-LA
0 20 40 60 80 100 120 140
Maximum Deviator Stress, psi (at 15 psi confining stress)
* Test not required for conditions representing significance level 1
Figure 5.20. Performance potential of recycled materials based on static

triaxial test.
Virgin (DGBL#1)
Virgin (DGBL#2)
RCP-IL-LS (50) (DGBL)
RCP-GV-LA (DGBL)
RAP-GR-CO (DGBL#2)
RCP-IL-LS (DGBL)
RCP-GR-SC (50) (DGBL)
Virgin OGDL
RCP-GR-SC (DGBL)
RAP-GR-CO (50) (DGBL#2)
RAP-GR-CO (OGDL)
RCP-SC-GR (50) (OGDL)

RAP-LS-MS (50) (OGDL)
RCP-GV-LA (50) (DGBL)

RAP-LS-MS (OGDL)
RAP-GR-CO (50) (OGDL)

RCP-SC-GR (OGDL)
RAP-GV-LA (50) (OGDL)
RAP-GV-LA (OGDL)
70 80 90 100 110 120 130 140 150 160 170 180 190
Failure Deviator Stress, psi
Figure 5.21. Performance potential based on repeated load triaxial test (OMC).
50
Significance Level 2 3 4
RCP-IL-LS (DGBL)
Virgin (DGBL#1)
RAP-GR-CO (DGBL#2)
RAP-GR-CO (50) (DGBL#2)
RCP-GR-SC (DGBL)
RCP-GR-SC (50) (DGBL)
RCP-GV-LA (50) (DGBL)
RCP-IL-LS (50) (DGBL)
Virgin (DGBL#2)
RCP-GV-LA (DGBL)
100 110 120 130 140 150 160 170 180 190
Failure Deviator Stress, psi
Figure 5.22. Performance potential based on repeated load triaxial test (saturated).
Virgin DGBL Blend#1

Virgin OGDL Blend
Virgin DGBL Blend#2
50/50 RAP-LS-MS
RCP-GV-LA
RAP-LS-MS
50/50 RAP-GR-CO
50/50 RAP-GV-LA
RAP-GR-CO OGDL reblend
50/50 RCP-GV-LA
50/50 RAP-GR-CO OGDL reblend
RCP-LS-IL
RAP-GR-CO
50/50 RCP-LS-IL
RAP-GV-LA
50/50 RCP-GR-SC
RCP-GR-SC
0 10 20 30 40 50 60 70
Resilient Modulus (stiffness) at 100 psi Bulk Stress, ksi
Figure 5.23. RAP and RCP performance potential based on stiffness (at OMC).
51
CHAPTER 6
Conclusions and Suggested Research
Conclusions mended for the use of RAP and RCP materials (or blends with
virgin aggregate) in different climatic conditions and traffic
Based on the results of the research in this project, the fol- levels.
lowing conclusions are made. The recommended aggregate tests and the ranges of test
parameters were based on laboratory test results. Further re-
1. Fatigue cracking, rutting/corrugations, depressions and search is needed to confirm the validity of these tests and
frost heave in flexible pavements and cracking, pumping/ ranges under service conditions.
faulting, frost heave, and erosion in rigid pavements are A field validation plan is suggested to further validate the
distresses associated with poor performance of the recycled suitability of using the performance-related tests identified in
aggregates used in the unbound layers of these pavements. this research as predictors of performance. The plan makes use
2. Properties of recycled aggregates used in unbound base of accelerated pavement testing (APT) and in-service pavement
and subbase pavement layers that affect pavement per- test sections.
formance include shear strength, stiffness, toughness, dura-
bility, frost susceptibility, and permeability. Shear strength
and stiffness (resilient modulus) have a much greater influ- Accelerated Pavement Testing
ence on the performance of an unbound aggregate layer Many state DOTs use accelerated pavement testing to eval-
than the other properties. uate potential construction materials, pavement designs, and
3. The following tests relate to the performance of recycled other pavement-related features. During APT, wheel loads are
materials used in unbound pavement layers: applied to specially constructed or in-service pavements to
• Screening tests for sieve analysis and the moisture-density
determine pavement response and performance under a con-
relationship, trolled and accelerated accumulation of damage in a short
• The Micro-Deval test for toughness,
time. It is recommended to construct flexible and rigid pave-
• Resilient modulus for stiffness,
ment sections with unbound recycled aggregate layers to eval-
• Static triaxial and repeated load at OMC and saturated
uate the merits of the reported research findings. By varying
for shear strength, and the characteristics of recycled aggregates used for unbound
• The tube suction test for frost susceptibility.
pavement layers, the effects of various aggregate properties on
the performance of unbound pavement layers can be assessed.
This field performance can then be compared with the per-
Suggested Research
formance predicted using the methodologies developed in this
Based on the work performed in this project, modifications research.
of the repeated load and moisture content tests are recom- The primary advantage of this approach is that the factors
mended. The modification of the repeated load test will im- that affect pavement performance could be more closely con-
prove the application of the seating load and reduce the inter- trolled. This is particularly important if the test involves study-
laboratory variability. The modification of the moisture content ing the effect of a single or a group of factors on pavement per-
test will allow air-drying at ambient temperature using a fan formance. The disadvantage of this approach is that long-term
blowing across loosely spread aggregate placed on a tarp. In strength loss due to poor durability and frost effects cannot be
addition, ranges for selected test parameters were recom- fully assessed.
52
In-Service Test Pavements and compared with test results from current DOT evaluation
procedures. Construction of the pavement project would be
Testing in-service pavements is proposed to evaluate followed to document construction practices; performance
the recommended procedures in actual practice. This ap- of the test pavement would be monitored for future analysis.
proach assesses the adaptability of the recommended tests There are benefits in testing in-service pavements, but there
to state DOTs’ current methods of evaluating recycled ag- are also major disadvantages. Testing the in-service pavement
gregates and compares the test results. In this study, the can provide the comfort of knowing that the pavement is “real”
performance of pavements incorporating unbound recy- in all respects. However, in most cases, in-service pavements do
cled material layers will be used to evaluate performance not allow for good control of the factors that may affect pave-
prediction accuracy. ment performance. Testing in-service pavement also limits, or
The study would involve identification of pavement proj- at least makes more difficult, the use of instrumentation in-
ects currently being designed that represent a range in traffic stalled in the pavement structure to measure response and per-
and climatic conditions. The recycled aggregate used in each formance. Also, it will generally take several years before per-
project would be tested using the recommended procedures formance data become available.
53
References
1. Saeed, A., Hall, J. W., Jr. and Barker, W., “Performance-Related Record No. 993, Transportation Research Board, National Research
Tests of Aggregates for Use in Unbound Pavement Layers.” NCHRP Council, Washington, DC (1984) pp. 28–35.
Report 453, Transportation Research Board, National Research 9. Thompson, M. R., “Important Properties of Base and Subgrade
Council, Washington, DC (2001) 55 pp. Materials.” Conference on Crushed Stone for Road and Street Con-
2. National Stone Association, The Aggregate Handbook. National Stone struction and Reconstruction, National Crushed Stone Associa-
Association, Washington, DC (1991). tion, Proceedings (1984).
3. Barksdale, R. D. and Itani, S. Y., “Influence of Aggregate Shape on 10. Chini, A. and Kuo, S. S., “Guidelines and Specifications for the Use
Base Behavior.” Transportation Research Record No. 1227, Trans- of Reclaimed Aggregates in Pavement.” Florida Department of
portation Research Board, National Research Council, Washing- Transportation Final Report for Contract BA 509, State Materials Of-
ton, DC (1989) pp. 173–182. fice, Gainesville, FL (1998).
4. Lees, G., “The Measurement of Particle Shape and Its Influence in 11. Rathje E. M., Rauch, A. F., Folliard, K. J., Viyanant, C., Ogalla, M.,
Engineering Materials.” Journal of the British Granite and Whin- Trejo, D., Little, D. and Esfeller, M, “Recycled Asphalt Pavement
stone Federation, Vol. 4, No 2, London, United Kingdom (1964). and Crushed Concrete Backfill: Results from Initial Durability and
5. Thompson, M. R., “Factors Influencing the Field Stability of Soil- Geotechnical Tests.” Center for Transportation Research Report
Aggregate Mixtures.” Materials Research Standards, Vol 7, No 1, 4177-2, The University of Texas at Austin, Texas (2002) 70 pp.
American Society of Testing and Materials, Philadelphia, PA (1987). 12. Saarenketo, T., and Scullion, T. Using Electrical Properties to
6. Kuo, S. S, Mahgoub, H. S. and Ortega, J. E., “Use of Recycled Con- Classify the Strength Properties of Base Course Aggregates, Re-
crete Made with Florida Limestone Aggregate for a Base Course in search Report 1341-2. Project No. 1341. Texas Transportation In-
Flexible Pavement.” Florida Department of Transportation Final stitute, College Station, TX. July 1995.
Report for Contract BC 409, State Materials Office, Gainesville, FL 13. Rangarajy, P. R., Edlinski, J., and Amirkhanian, S. Evaluation of
(2001) 212 pp. South Carolina Aggregate Durability Properties. Report No.
7. Liu, S. J., Leyapalan, J. K. and Lytton, R. L., “Characteristics of Base FHWA-SC-05-01, South Carolina Department of Transportation,
and Subgrade Drainage of Pavements.” Transportation Research Columbia, SC, January 2005.
Record No. 945, Transportation Research Board, National Research 14. American Association of State Highway and Transportation Officials
Council, Washington, DC (1983) pp. 1–10. (AASHTO), Guide for Design of Pavement Structures, Washing-
8. Liu, S. J. and Lytton, R. L., “Rainfall Infiltration, Drainage, and ton, DC, 1993.
Load-Carrying Capacity of Pavements.” Transportation Research
Abbreviations and acronyms used without definitions in TRB publications:
AAAE American Association of Airport Executives
AASHO American Association of State Highway Officials
AASHTO American Association of State Highway and Transportation Officials
ACI–NA Airports Council International–North America
ACRP Airport Cooperative Research Program
ADA Americans with Disabilities Act
APTA American Public Transportation Association
ASCE American Society of Civil Engineers
ASME American Society of Mechanical Engineers
ASTM American Society for Testing and Materials
ATA Air Transport Association
ATA American Trucking Associations
CTAA Community Transportation Association of America
CTBSSP Commercial Truck and Bus Safety Synthesis Program
DHS Department of Homeland Security
DOE Department of Energy
EPA Environmental Protection Agency
FAA Federal Aviation Administration
FHWA Federal Highway Administration
FMCSA Federal Motor Carrier Safety Administration
FRA Federal Railroad Administration
FTA Federal Transit Administration
IEEE Institute of Electrical and Electronics Engineers
ISTEA Intermodal Surface Transportation Efficiency Act of 1991
ITE Institute of Transportation Engineers
NASA National Aeronautics and Space Administration
NASAO National Association of State Aviation Officials
NCFRP National Cooperative Freight Research Program
NCHRP National Cooperative Highway Research Program
NHTSA National Highway Traffic Safety Administration
NTSB National Transportation Safety Board
SAE Society of Automotive Engineers
SAFETEA-LU Safe, Accountable, Flexible, Efficient Transportation Equity Act:
A Legacy for Users (2005)
TCRP Transit Cooperative Research Program
TEA-21 Transportation Equity Act for the 21st Century (1998)
TRB Transportation Research Board
TSA Transportation Security Administration
U.S.DOT United States Department of Transportation

NCHRP RPT 598

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NCHRP RPT 598

Uploaded by

Copyright:

Available Formats

NATIONAL

NCHRP REPORT 598

*Membership as of January 2008.

NCHRP REPORT 598

TRANSPORTATION RESEARCH BOARD

Published reports of the

NATIONAL COOPERATIVE HIGHWAY RESEARCH PROGRAM

Transportation Research Board

and can be ordered through the Internet at:

CRP STAFF FOR NCHRP REPORT 598

NCHRP PROJECT 4-31

This report contains recommendations for performance-related procedures to test and

Appendix A: Literature Review and Background Information

Background and Research Approach

Project Background Task 2. Identify the performance parameters of pavements

Research Approach Report Organization

Pavement Performance and Recycled

• A working platform for construction, Rigid Pavements

Table 2.1. Flexible pavement performance parameters and contributing factors.

Distress Description of distress Unbound layer failure mechanism Contributing factors

foundation soil or consolidation load applications. Changes in base material combined

Table 2.2. Rigid pavement performance parameters and contributing factors.

Distress Description of distress Unbound layer failure mechanism Contributing factors

crack and faulting at the crack or develops, increased load is placed

Table 2.3. Recycled aggregate particle properties that

Physical properties Chemical properties Mechanical properties

Table 2.4. Relevance of recycled material mass properties for various

Relevance of Mass Property to the Use of Recycled Material as

Table 2.5. Links between aggregate properties and performance.

Pavement Performance Related aggregate

Toughness Particle strength, particle degradation, gradation

Selection of Candidate Test Methods

Property Performance Practi- Com-

Table 3.2. Tests selected for the laboratory test program.

Table 3.3. Stress control for repeated load triaxial test.

Confining Contact Stress Cyclic Stress

Figure 3.1. Load pulse for repeated load triaxial tests.

Table 3.4. Selected materials and expected performance potential.

Table 3.5. Gradation for Illinois DOT CA 6 coarse

Sieve Size Percent passing (percent finer)

Table 3.6. Louisiana DOT&D gradation for

Sieve Size Percent passing (percent finer)

of I-20 near Monroe, Louisiana. The in-place PCC pavement

Sieve Size Opening (inch)

Figure 3.4. RCP-GV-LA as-received and typical Louisiana DOT&D

Table 3.7. Gradation requirements for

Sieve Size Percent passing (percent finer)

in-place HMA surface was in good condition before milling;

RAP with Granite RAP with Gravel

Figure 3.6. RCP-GR-SC as-received gradation and typical SC DGBL gradation.

Figure 3.8. RAP-LS-MS as-received and Mississippi DOT

Figure 3.9. Photograph of RAP-LS-MS material

Typical NSA DGBL gradation

Figure 3.10. RAP-GR-CO as-received gradation and typical NSA DGBL

Figure 3.12. RAP-GV-LA as-received and typical NSA OGDL gradation.

Figure 3.13. Photograph of RAP-GV-LA material used

Laboratory Test Program and Test Results

Table 4.1. Percent passing for laboratory testing.

Sieve size DGBL OGDL

Table 4.2. OMC and density data.

Maximum Dry OMC

Figure 4.1. Gradation results of RCP as-received DGBL samples.