PHD Thesis Jsmith

Recycled Concrete Aggregate A Viable
Aggregate Source For Concrete Pavements
by
James Trevor Smith
A thesis
presented to the University of Waterloo
in fulfillment of the
thesis requirement for the degree of
Doctor of Philosophy
in
Civil Engineering
Waterloo, Ontario, Canada, 2009
James Trevor Smith 2009

Authors Declaration
I hereby declare that I am the sole author of this thesis. This is a true copy of the thesis, including any
required final revisions, as accepted by my examiners.
I understand that my thesis may be made electronically available to the public.
ii
Abstract
Virgin aggregate is being used faster than it is being made available creating a foreseeable shortage in
the future. Despite this trend, the availability of demolished concrete for use as recycled concrete
aggregate (RCA) is increasing. Using this waste concrete as RCA conserves virgin aggregate, reduces
the impact on landfills, decreases energy consumption and can provide cost savings. However, there
are still many unanswered questions on the beneficial use of RCA in concrete pavements.
This research addresses the many technical and cost-effective concerns regarding the use of RCA
in concrete pavements by identifying concrete mixture and proportioning designs suitable for jointed
plain concrete pavements; constructing test sections using varying amounts of RCA; monitoring
performance through testing, condition surveys and sensor data; modeling RCA pavement
performance; and predicting life cycle costs.
The research was carried out as a partnership between the Centre for Pavement and Transportation
Technology (CPATT) at the University of Waterloo, the Cement Association of Canada, Dufferin
Construction, and the Natural Sciences and Engineering Research Council of Canada.
The literature review provides an overview of sustainability and key performance indicators, the
material properties of RCA both as an aggregate and in concrete, concrete mixture and proportioning
designs with RCA, performance of existing RCA pavements, and the implementation of RCA
highlighting some examples where RCA has been used successfully.
Twelve preliminary mixes were developed using three total cementitious contents amounts of 315
kg/m3, 330 kg/m3, and 345 kg/m3 to determine four suitable mixes with varying coarse RCA contents
(0%, 15%, 30% and 50%) to place at the CPATT test track. At 28-days, all of the twelve mixes
exceed the 30 MPa design strength.
Four test sections containing 0%, 15%, 30% and 50% coarse RCA were constructed in June 2007.
The test sections had identical cross sections consisting of 250 mm portland cement concrete
(PCC), 100 mm asphalt-stabilized OGDL and a 450 mm granular base. For each coarse RCA
content, one slab was instrumented with six vibrating wire concrete embedment strain gages to
measure long-term longitudinal and transverse strain due to environmental changes, two vibrating
wire vertical extensometers to monitor slab curling and warping, two vibrating wire inter-panel
extensometers to monitor joint movement, and two maturity meters to measure maturity and
temperature.
Quality assurance and quality control (QA/QC) testing showed that the mixes containing RCA
exhibited similar or improved performance when compared to the conventional concrete for
compressive and flexural strength, freeze-thaw durability and coefficient of thermal expansion.
Pavement performance of the four test sections was evaluated using visual surveys following the
Ontario Ministry of Transportations Manual for Condition rating of Rigid Pavements. Nine
pavement evaluations have been performed every two to four months since construction. All test
sections are in excellent condition with pavement condition index (PCI) values greater than 85 after
two years in-service and approximately three hundred thousand Equivalent Single Axle Loads.
iii
Sensor data from the strain gauges, and vertical and inter-panel extensometers are providing
consistent results between the test sections.
Long-term performance modeling using the Mechanistic-Empirical Pavement Design Guide (ME-
PDG) showed improved performance with respect to cracked slabs, joint faulting, and pavement
roughness as the RCA content increased. Multivariable sensitivity analysis showed that the
performance results were sensitive to CTE, unit weight, joint spacing, edge support, surface
absorption, and dowel bar diameter.
Life cycle cost analysis (LCCA) illustrated the savings that can be expected using RCA as a
replacement aggregate source as the cost of virgin aggregate increase as the sources becomes
depleted. Multivariable sensitivity analysis showed that the LCCA results were sensitive to
construction costs, discount rate, and maintenance and rehabilitation quantities.
iv
Acknowledgements
I wish to express a sincere thank you to my supervisor Dr. Susan Tighe. Your support, guidance, and
encouragement are greatly appreciated.
I would also like to acknowledge the Centre for Pavement and Transportation Technologies
(CPATT) at the University of Waterloo, the Cement Association of Canada (CAC), Dufferin
Construction, Steed and Evans, Smith Recycling, the Natural Sciences and Engineering Research
Council of Canada (NSERC), the Regional Municipality of Waterloo, Environment Canada, and
Stantec Consulting Ltd. for their financial support and contributions to this research.
Special thank you to my wife Jennifer, and our children Victoria, Benjamin, and Nathaniel for their
support and encouragement during this process.
v
Table of Contents
List of Figures.............................................................................................................................................. xi
List of Tables .............................................................................................................................................xvi
Chapter 1 Introduction ................................................................................................................................. 1
1.1 Background ........................................................................................................................................ 1
1.1.1 Current Aggregate Shortage ...................................................................................................... 1
1.1.2 Impacts of Waste Concrete ........................................................................................................ 1
1.1.3 Costs ............................................................................................................................................ 2
1.1.4 Achieving Sustainability with Recycled Aggregate Concrete................................................. 2
1.1.5 Barriers to Recycled Concrete Aggregate Use ......................................................................... 2
1.2 Thesis Scope and Objectives............................................................................................................. 3
1.3 Project Significance ........................................................................................................................... 4
Chapter 2 Literature Review........................................................................................................................ 5
2.1 Sustainability...................................................................................................................................... 5
2.2 Recycled Concrete Aggregate........................................................................................................... 5
2.2.1 Producing Recycled Concrete Aggregate ................................................................................. 5
2.2.2 Current Use of Recycled Concrete Aggregate.......................................................................... 7
2.2.3 Properties of RCA ...................................................................................................................... 8
2.2.4 Properties of RCA Concrete ...................................................................................................... 9
2.2.5 RCA Properties Summary........................................................................................................ 17
2.3 RCA Concrete Mix Designs............................................................................................................ 19
2.3.1 Components of concrete mixes................................................................................................ 19
2.3.2 Mixing Methods ....................................................................................................................... 20
2.3.3 Testing....................................................................................................................................... 20
2.3.4 Mix Designs with RCA ............................................................................................................ 22
2.3.5 Pavement Performance............................................................................................................. 23
2.4 Implementation of RCA Use Steps in the Right Direction Going Green .......................... 23
2.4.1 FHWA Recycled Materials Policy .......................................................................................... 23
2.4.2 LEED Program ......................................................................................................................... 24
2.4.3 Success Stories.......................................................................................................................... 24
Chapter 3 Materials and Test Mixes ......................................................................................................... 26
3.1 Aggregate Sources ........................................................................................................................... 26
vi
3.1.1 Sources of Virgin Coarse Aggregate .......................................................................................26
3.1.2 Sources of Recycled Coarse Aggregate ..................................................................................26
3.1.3 Source of Virgin Fine Aggregate.............................................................................................27
3.2 Aggregate Testing ............................................................................................................................27
3.2.1 MTO Guidelines for Aggregate ...............................................................................................27
3.2.2 Aggregate Test Results.............................................................................................................28
3.3 Cement and Slag ..............................................................................................................................33
3.4 Air Entrainment................................................................................................................................34
3.5 Water Reducer..................................................................................................................................34
3.6 Open Graded Drainage Layer .........................................................................................................34
3.7 Concrete Mixture Design ................................................................................................................35
3.8 Concrete Mixture Proportioning .....................................................................................................36
3.8.1 Proportioning Example.............................................................................................................36
3.9 Testing Method ................................................................................................................................39
3.9.1 Batching Method ......................................................................................................................41
3.10 Results ............................................................................................................................................41
3.10.1 Plastic Properties ....................................................................................................................41
3.10.2 Hardened Properties ...............................................................................................................42
3.10.3 Selection of Mix .....................................................................................................................45
3.11 Statistical Analysis.........................................................................................................................45
Chapter 4 Pavement Design.......................................................................................................................47
4.1 Soil Investigation .............................................................................................................................47
4.2 FWD Testing ....................................................................................................................................48
4.3 Site Survey........................................................................................................................................50
4.4 Estimated Traffic Load ....................................................................................................................51
4.5 Sensors ..............................................................................................................................................52
4.6 Design Thickness .............................................................................................................................55
4.7 Preconstruction Meetings ................................................................................................................55
Chapter 5 Construction ..............................................................................................................................56
5.1 Granular Base ...................................................................................................................................56
5.2 Sensor Conduit .................................................................................................................................56
5.3 Asphalt Stabilized Open Graded Drainage layer ...........................................................................57
vii
5.4 Jointed Plain Concrete Pavement ................................................................................................... 59
5.5 Early Age Behaviour ....................................................................................................................... 65
5.6 Joints................................................................................................................................................. 67
5.7 Sensors.............................................................................................................................................. 69
Chapter 6 Test Results ............................................................................................................................... 74
6.1 Thickness.......................................................................................................................................... 74
6.2 MIT Scan-2 ...................................................................................................................................... 74
6.3 Compressive Strength...................................................................................................................... 77
6.4 Flexural Strength.............................................................................................................................. 78
6.5 Maturity ............................................................................................................................................ 80
6.6 Freeze-Thaw..................................................................................................................................... 82
6.7 Coefficient of Thermal Expansion.................................................................................................. 85
6.8 Testing Summary ............................................................................................................................. 88
Chapter 7 Field Performance..................................................................................................................... 89
7.1 Method.............................................................................................................................................. 89
7.2 Evaluations ....................................................................................................................................... 91
7.2.1 Evaluation #1 June 18, 2007................................................................................................. 91
7.2.2 Evaluation #2 September 20, 2007 ...................................................................................... 92
7.2.3 Evaluation #3 November 9, 2007......................................................................................... 92
7.2.4 Evaluation #4 April 4, 2008.................................................................................................. 93
7.2.5 Evaluation #5 June 25, 2008................................................................................................. 96
7.2.6 Evaluation #6 September 21, 2008 ...................................................................................... 96
7.2.7 Evaluation #7 Nov 1, 2008 ................................................................................................... 98
7.2.8 Evaluation #8 April 23, 2009................................................................................................ 99
7.2.9 Evaluation #9 June 6, 2009................................................................................................. 100
7.3 Summary ........................................................................................................................................ 102
Chapter 8 Sensor Evaluation ................................................................................................................... 103
8.1 Datalogger Program....................................................................................................................... 104
8.2 Sensor Readings............................................................................................................................. 104
8.2.1 Strain Gauges .......................................................................................................................... 104
8.2.2 Vertical Extensometers .......................................................................................................... 107
8.2.3 Inter-Panel Extensometers ..................................................................................................... 109
viii
8.3 Predictive Models ..........................................................................................................................113
8.3.1 Temperature ............................................................................................................................113
8.3.2 Strain........................................................................................................................................114
8.3.3 Vertical Extensometer ............................................................................................................117
8.3.4 Inter-Panel Extensometer .......................................................................................................118
Chapter 9 Mechanistic-Empirical Pavement Design Guide Performance............................................120
9.1 ME-PDG Background and How It Works....................................................................................120
9.2 ME-PDG Sensitivity ......................................................................................................................123
9.2.1 Experimental Design ..............................................................................................................125
9.2.2 Results .....................................................................................................................................128
9.3 Modeling Test Sections .................................................................................................................133
9.3.1 Materials..................................................................................................................................133
9.3.2 Climate ....................................................................................................................................135
9.3.3 Test Section Results and Discussion .....................................................................................136
Chapter 10 Life Cycle Cost Analysis......................................................................................................140
10.1 LCCA Steps..................................................................................................................................140
10.1.1 Step 1: Establish Alternative Designs .................................................................................140
10.1.2 Step 2: Determine Performance Periods and Activity Timings.........................................140
10.1.3 Step 3: Estimate Agency and User Costs............................................................................141
10.1.4 Step 4: Compute Life Cycle Costs ......................................................................................143
10.1.5 Step 5: LCCA Results Scenario 1 Higher Costs for RCA Concretes ............................144
10.1.6 Step 5: LCCA Results Scenario 2 Equal Costs for RCA and Conventional Concretes 146
10.1.7 Step 5: LCCA Results Scenario 3 Higher Costs for Conventional Concretes ..............148
10.2 Sensitivity.....................................................................................................................................150
10.2.1 Convergence .........................................................................................................................150
10.2.2 Input Parameters ...................................................................................................................150
10.2.3 Discount Rate........................................................................................................................152
10.3 Discussion ....................................................................................................................................153
Chapter 11 Conclusions and Recommendations ....................................................................................154
11.1 Conclusions ..................................................................................................................................154
11.2 Recommendations........................................................................................................................157
Bibliography .............................................................................................................................................158
ix
Appendices
Appendix A Design Tables...................................................................................................................... 168
Appendix B Preconstruction Meeting Summaries ................................................................................. 173
Appendix C Pavement Distress ............................................................................................................... 183
Appendix D Datalogger Program............................................................................................................ 187
Appendix E Sensor Results ..................................................................................................................... 193
x
List of Figures
Figure 1.1 Research Framework.................................................................................................................. 4
Figure 2.1 Closed-Loop Concrete System .................................................................................................. 6
Figure 2.2 Cone and Jaw Compression Crushers ....................................................................................... 6
Figure 2.3 Vertical and Horizontal Impact Crushers ................................................................................. 6
Figure 2.4 United States Using RCA as Base Material ............................................................................. 7
Figure 2.5 United States Using RCA as Concrete Aggregate ................................................................... 7
Figure 3.1 Waste Concrete and Recycled Aggregate...............................................................................26
Figure 3.2 Materials Finer than 75m Test Results .................................................................................29
Figure 3.3 Absorption Test Results...........................................................................................................29
Figure 3.4 Bulk Relative Density Test Results.........................................................................................30
Figure 3.5 Crushed Particles Test Results ................................................................................................30
Figure 3.6 Flat and Elongated Test Results ..............................................................................................30
Figure 3.7 Unconfined Freeze-Thaw Loss Test Results ..........................................................................31
Figure 3.8 Petrographic Number Test Results..........................................................................................31
Figure 3.9 Micro-Deval Abrasion Loss Test Results...............................................................................31
Figure 3.10 Organic Impurities Test Results............................................................................................32
Figure 3.11 Combined 37.5 mm and 19.0 mm Virgin Coarse Aggregate Gradation ............................32
Figure 3.12 19.0 mm minus RCA Coarse Aggregate Gradation.............................................................32
Figure 3.13 Fine Aggregate Gradation......................................................................................................33
Figure 3.14 Compressive Strength Development of Neat Paste .............................................................33
Figure 3.15 HL8 Gradation for Asphalt Stabilized Open Graded Drainage layer .................................34
Figure 3.16 Waterloo Climograph.............................................................................................................35
Figure 3.17 Waste Concrete and Recycled Aggregate Combined Gradation ........................................40
Figure 3.18 Compressive Strength Development 0% Coarse RCA........................................................43
Figure 3.19 Compressive Strength Development 15% Coarse RCA......................................................44
Figure 3.22 Compressive Strength at 28 Days .........................................................................................45
Figure 4.1 Dynatest HWD .........................................................................................................................48
Figure 4.2 Test Section Profile ..................................................................................................................50
Figure 4.3 Monte Carlo Simulation Results .............................................................................................52
xi
Figure 4.4 Sensor Conduit ......................................................................................................................... 54
Figure 4.5 Revised Sensor Locations........................................................................................................ 54
Figure 5.1 Digging Sensor Conduit Trenches .......................................................................................... 56
Figure 5.2 Marshall Hammer Compaction Around Sensor Conduit....................................................... 57
Figure 5.3 Paving and Compaction of ODGL.......................................................................................... 58
Figure 5.4 Pavement Construction............................................................................................................ 59
Figure 5.5 Skewing of the Dowel Bars ..................................................................................................... 61
Figure 5.6 Condition of JPCP.................................................................................................................... 62
Figure 5.7 Placing Concrete for Removal ................................................................................................ 62
Figure 5.8 Removal of Waste Concrete.................................................................................................... 63
Figure 5.9 Reestablishing Joints Prior to Paving ..................................................................................... 63
Figure 5.10 HIPERPAV II Results 0% Coarse RCA, Day 1 .................................................................. 65
Figure 5.11 HIPERPAV II Results 15% Coarse RCA, Day 1 ................................................................ 66
Figure 5.15 Joint Offsets............................................................................................................................ 67
Figure 5.16 Initial Saw Cuts ...................................................................................................................... 68
Figure 5.17 Widening and Sealing the Initial Saw Cut ........................................................................... 69
Figure 5.18 Pavement Damage ................................................................................................................. 69
Figure 5.19 Strain Gauge and Maturity Setup.......................................................................................... 70
Figure 5.20 Vertical Extensometer ........................................................................................................... 70
Figure 5.21 Mounding the Vertical Extensometer with Concrete .......................................................... 71
Figure 5.22 Mounding the Stain Gauges and Maturity Sensors with Concrete ..................................... 71
Figure 5.23 Inter-panel Extensometer Blockout Before Paving ............................................................. 71
Figure 5.24 Location of Multiplexer and Dataloger Cabinets................................................................. 72
Figure 5.25 Relocated Multiplexers and Datalogger, and Wiring .......................................................... 73
Figure 6.1 Thickness of JPCP ................................................................................................................... 74
Figure 6.2 MIT Scan-2............................................................................................................................... 75
Figure 6.3 Distribution of Dowel Bar Misalignment............................................................................... 75
Figure 6.4 Joint Scores............................................................................................................................... 76
Figure 6.5 Rolling Average Joint Scores .................................................................................................. 77
xii
Figure 6.6 Compressive Strength Results.................................................................................................78
Figure 6.7 Flexural Strength Testing.........................................................................................................79
Figure 6.8 Flexural Strength Testing Results ...........................................................................................80
Figure 6.9 RCA Maturity Curves ..............................................................................................................81
Figure 6.10 Beams in the Freeze-Thaw Bed.............................................................................................82
Figure 6.11 Erudite MkIV Testing Equipment and Setup .......................................................................82
Figure 6.12 Changes in Relative Dynamic Modulus ...............................................................................84
Figure 6.13 CTE Values for RCA .............................................................................................................87
Figure 7.1 0% Coarse RCA Northbound View and 50% Coarse RCA Southbound View ...................91
Figure 7.2 Pavement Surface Tining Close-up.........................................................................................91
Figure 7.3 Joint Spalling ............................................................................................................................92
Figure 7.4 Pavement Edge Spalling and Shoulder Settlement ................................................................93
Figure 7.5 Evaluation 4 - Pavement Distresses ........................................................................................94
Figure 7.6 Core Hole Settlement ...............................................................................................................94
Figure 7.7 Transverse Depression from Slipform Paver..........................................................................95
Figure 7.8 Asphalt Adjoining Test Sections.............................................................................................95
Figure 7.9 Blockout Failure .......................................................................................................................95
Figure 7.10 Pavement Edge Spalling and Shoulder Settlement ..............................................................96
Figure 7.11 Joint Spalling ..........................................................................................................................97
Figure 7.12 New Raveling .........................................................................................................................97
Figure 7.13 Asphalt Cracking....................................................................................................................98
Figure 7.14 Pavement Edge Spalling and Shoulder Setlement ...............................................................99
Figure 7.15 Patching of Adjoining Asphalt ............................................................................................100
Figure 7.16 Evaluation 9 Test Section Conditions..............................................................................101
Figure 7.17 New Pothole in 0% Coarse RCA ........................................................................................101
Figure 7.18 Regraded Southbound Shoulder..........................................................................................102
Figure 7.19 PCI Progression...................................................................................................................102
Figure 8.1 Sample Raw Strain Data, 15% Coarse RCA ........................................................................105
Figure 8.2 Calculated Strain Profile, 15% Coarse RCA ........................................................................106
Figure 8.3 Slab Curling............................................................................................................................107
Figure 8.4 Sample Vertical Displacement, 15% Coarse RCA ..............................................................108
Figure 8.5 Normalized Vertical Displacement, 15% Coarse RCA .......................................................108
xiii
Figure 8.6 Sample Horizontal Displacement, 15% Coarse RCA.......................................................... 110
Figure 8.7 Normalized Horizontal Displacement, 15% Coarse RCA................................................... 110
Figure 8.8 Calculated and Measured Joint Movement .......................................................................... 112
Figure 8.9 Strain Temperature Prediction .............................................................................................. 113
Figure 8.10 Actual vs. Predicted Raw Transverse Strain, Centre Slab, Upper .................................... 115
Figure 8.11 Actual vs. Predicted Raw Transverse Strain, Centre Slab, Lower.................................... 115
Figure 8.12 Actual vs. Predicted Vertical Displacement at Edge of Pavement ................................... 117
Figure 8.13 Actual vs. Predicted Vertical Displacement at Centreline ................................................ 117
Figure 8.14 Actual vs. Predicted Horizontal Displacement at Edge of Pavement............................... 118
Figure 8.15 Actual vs. Predicted Horizontal Displacement at Centreline ............................................ 119
Figure 9.1 ME-PDG Flow Chart ............................................................................................................. 120
Figure 9.2 Example Results..................................................................................................................... 128
Figure 9.3 JPCP Crack Effect Results .................................................................................................... 130
Figure 9.4 JPCP Faulting Effect Results ................................................................................................ 131
Figure 9.5 JPCP Pavement Roughness Effect Results........................................................................... 131
Figure 9.6 Design Crack Effect Results.................................................................................................. 131
Figure 9.7 Design Faulting Effect Results.............................................................................................. 132
Figure 9.8 Design Pavement Roughness Effect Results........................................................................ 132
Figure 9.9 Precipitation Distribution Scenarios ..................................................................................... 135
Figure 9.10 Percent Slabs Cracked Performance Results...................................................................... 136
Figure 9.11 Faulting Performance Results ............................................................................................. 137
Figure 9.12 Pavement Roughness Performance Results ....................................................................... 137
Figure 10.1 Standard Rigid LCCA Model Used .................................................................................... 141
Figure 10.2 Triangle Distribution with Five Iterations.......................................................................... 143
Figure 10.3 Cost Associated with Structural and Functional Deterioration......................................... 144
Figure 10.4 Combined Costs ................................................................................................................... 144
Figure 10.5 Results of LCCA Scenario 1 ............................................................................................... 145
Figure 11.6 Cumulative Distribution Probabilities Scenario 1 ............................................................. 146
Figure 10.8 Cumulative Distribution Probabilities Scenario 2 ............................................................. 147
Figure 10.10 Cumulative Distribution Probabilities Scenario 3 ........................................................... 149
xiv
Figure 10.11 Convergence Results..........................................................................................................150
Figure 11.12 Significant Inputs ...............................................................................................................152
Figure 10.13 Impact of Discount Rate on Present Worth......................................................................153
xv
List of Tables
Table 2.1 RCA Material Properties Summary ......................................................................................... 18
Table 2.2 RCA Concrete Properties Summary ........................................................................................ 18
Table 2.3 Summary of RCA Specifications ............................................................................................. 21
Table 2.4 Maximum Limit of Harmful Substances Allowed in RCA Concrete .................................... 21
Table 2.5 RCA Mix Designs ..................................................................................................................... 22
Table 3.1 Coarse Aggregate Gradation Limits......................................................................................... 27
Table 3.2 Limits of Coarse Aggregate Properties .................................................................................... 27
Table 3.3 Fine Aggregate Gradation Limits............................................................................................. 28
Table 3.4 Limits of Fine Aggregate Properties ........................................................................................ 28
Table 3.5 Material Properties Used in Concrete Mixture Proportioning................................................ 36
Table 3.6 Mixture Proportioning Values .................................................................................................. 40
Table 3.7 Plastic Properties ....................................................................................................................... 42
Table 3.8 Air Content and Spacing Factor ............................................................................................... 42
Table 3.9 Results of ANOVA ................................................................................................................... 46
Table 4.1 Borehole Log OW7-83.............................................................................................................. 47
Table 4.4 Geophone Sensor Configuration (mm) .................................................................................... 49
Table 4.5 FWD Results.............................................................................................................................. 49
Table 4.6 Manufacturer Stated Axle Loads.............................................................................................. 52
Table 5.1 ODGL Load Details .................................................................................................................. 58
Table 5.2 Concrete Fresh Properties ......................................................................................................... 60
Table 5.3 Day 1 Paving.............................................................................................................................. 62
Table 5.4 Day 1 Texturization................................................................................................................... 63
Table 5.5 Day 2 Paving.............................................................................................................................. 64
Table 5.6 Day 2 Texturization................................................................................................................... 64
Table 6.1 Joint Score Weighting Factors.................................................................................................. 76
Table 6.2 Results of t-Ttest........................................................................................................................ 78
Table 6.3 Strength Estimates of Placed Concrete .................................................................................... 81
Table 6.4 Properties of Beams Prior to Testing ....................................................................................... 83
Table 6.5 Freeze-Thaw Testing Results ................................................................................................... 84
xvi
Table 6.6 CTE Testing Results..................................................................................................................86
Table 6.7 CTE Testing Results..................................................................................................................88
Table 7.1 Pavement Distress Weights.......................................................................................................90
Table 7.2 Distress Severity Levels and Weights ......................................................................................90
Table 7.3 Distress Density Levels and Weights .......................................................................................90
Table 7.4 PCI Rating Scale........................................................................................................................90
Table 7.5 Evaluation 3 Test Track Conditions......................................................................................92
Table 7.10 Evaluation 8 Test Track Conditions....................................................................................99
Table 7.11 Evaluation 9 Test Track Conditions..................................................................................100
Table 8.1 Maximum and Minimum Strain Values.................................................................................106
Table 8.2 Maximum and Minimum Vertical Displacement Values .....................................................109
Table 8.3 Maximum and Minimum Horizontal Displacement Values .................................................111
Table 8.4 Joint Movement Calculation Values ......................................................................................112
Table 8.5 Temperature Coefficient of Determination Values ...............................................................114
Table 8.6 Predictive Strain Equations.....................................................................................................116
Table 8.7 Predictive Vertical Displacement Equations..........................................................................118
Table 8.8 Predictive Horizontal Displacement Equations .....................................................................119
Table 9.1 MEPDG Sensitivity Analysis .................................................................................................124
Table 9.2 JPCP Layer Sensitivity Factors and Values ...........................................................................126
Table 9.3 Design Sensitivity Factors and Values ...................................................................................126
Table 9.4 PB Design Run Variability Levels .........................................................................................127
Table 9.5 JPCP Factor Results ................................................................................................................129
Table 9.6 Design Factor Results..............................................................................................................129
Table 9.7 ME-PDG Inputs for JPCP and Asphalt Layers......................................................................133
Table 9.8 ME-PDG Inputs for Granular Base and CL Subgrade ..........................................................134
Table 9.9 CTE Variability Results ..........................................................................................................138
Table 9.10 Summary Results...................................................................................................................139
Table 10.1 Proposed Activity Timing Schedules ...................................................................................141
xvii
Table 10.2 Material and Activity Costs .................................................................................................. 142
Table 10.3 Results of LCCA Scenario 1................................................................................................. 145
Table 10.6 Value of Factors in LCCA .................................................................................................... 150
Table 10.7 Present Worth Sensitivity Results ........................................................................................ 151
xviii
Chapter 1
Introduction
1.1 Background
1.1.1 Current Aggregate Shortage
There is a dramatic decline in good quality aggregate available for construction use. World wide
aggregate use is estimated to be ten to eleven billion tonnes each year. Of this, approximately eight
billion tonnes of aggregate (sand, gravel, and crushed rock) is being used in portland cement concrete
(PCC) every year [Naik 2005, Mehta 2001]. In central Canada, fourteen tonnes of aggregate are
consumed per person each year. However, for every three tonnes of aggregate that is produced only
one tonne is replaced by opening new aggregate sources or through recycling [McMacNaughton
2004]. Ontario is currently using aggregate faster than it is being made available resulting in an
aggregate shortage [APAO 2004].
The current state of aggregate resources in Ontario is not fully known since the last detailed study
was done in 1992 by the Ministry of Natural Resources (MNR) in State of the Resource Report
[Miller 2005]. From 1992 to 2003 Ontarios yearly consumption of aggregate was approximately 170
million tonnes. More than sixteen tonnes of aggregate are used per person in Ontario each year
[APAO 2004]. To construct one kilometer of six-lane expressway, 51,800 tonnes of aggregate are
used. Aggregate production in Ontario is currently produced by 2800 licensed pit and quarries and it
exceeded 160 million tonnes in 2001 [Miller 2005]. Aggregate consumption in Ontario over the next
25 years is estimated at four billion tonnes [APAO 2004]. Ontario is facing an aggregate shortage. It
is estimated that some urban areas will run out of aggregate by 2010 [Miller 2005].
1.1.2 Impacts of Waste Concrete
There is a critical shortage of natural aggregate and an increasing amount of demolished concrete
[Hansen 1984]. It is estimated that 150 million ton of concrete waste is produced in the United States
annually [Salem 2003]. Concrete structures that are designed to have service lives of at least 50 years
have to be demolished after 20 or 30 years because of early deterioration. In 2005, the American
Society of Civil Engineers reported US infrastructure in poor condition with an estimated repair cost
of $1.6 trillion over five years [ASCE 2005]. The environmental impact of waste concrete is
significant. Not only is there the environmental impact of transporting the waste concrete away from
the site but the waste concrete also fills up valuable space in landfills. The United States produces
123 million tons of waste from building demolition, and most ends up in landfills [FHWA 2004].
Construction and demolition (C&D) waste makes up a large portion of all generated solid waste
[Meyer 2008]. In 1980 the Environmental Resources Limited in the East European Communities
(EEC) estimated 80 million tonnes of demolition waste, mostly concrete, are produced each year.
This number is expected to double by 2000, and triple by 2020 [Bairagi 1990]. There is a huge
1
potential to reuse this material as source of new aggregate. Due to concerns with space and cost,
traditional disposal of C&D in landfills is no longer an acceptable option [Meyer 2008].
1.1.3 Costs
The cost of quality aggregate has increased above the inflation rate and it is projected that this trend
will continue as further restrictions are placed on this resource in the future [MacNaughton 2004].
Concrete C&D waste will be recycled if it is less expensive than disposing of it in a landfill and
recycled concrete aggregate (RCA) will be used if it is less expensive than virgin aggregate of similar
quality [Kamel 2008]. RCA use is based on economics, including the cost of transporting C&D
waste and virgin aggregate, the cost of C&D disposal, and government intervention on tipping fees
and mandatory usage through legislation [Meyer 2008]. Approximately 60% of aggregate cost is due
to transportation [APAO 2004]. The economics are starting to make recycled materials more
attractive [Turley 2003]. It is estimated that the Canadian ready mixed concrete industry could save
approximately $300 million annually by using RCA [NRMCA 2007].
1.1.4 Achieving Sustainability with Recycled Aggregate Concrete
Sustainability is defined as Meeting the needs of the present without compromising the ability of the
future generations to meet their own needs [Naik 2005]. The current usage of aggregate is not
sustainable as demonstrated by the growing shortage of natural aggregates in urban area. Recycling
concrete, from deteriorated concrete structures, would reduce the negative impact on the environment
and increase sustainability of aggregate resources [Oikonomou 2005, Hansen 1985]. Using RCA
conserves virgin aggregate, reduces the impact on landfills and decreases energy consumption
[FHWA 2004]. Using RCA, creates cost savings in the transportation of aggregate and waste
products, and in waste disposal [FHWA 2004, ACPA 1993]. It is estimated that using RCA can save
up to $4.80 m2 [ACPA 2009]. Finding ways to re use C&D waste and minimize things that are not
suitable for reuse will increase sustainability [Turley 2003].
1.1.5 Barriers to Recycled Concrete Aggregate Use
There are several barriers to overcome in order for RCA to become widely accepted [Rashwan 1997].
Initially there is a high investment cost to purchase concrete crushers. In addition, maintenance costs
of concrete crushers are significant.
In Ontario, non-virgin aggregate consumption is estimated at three percent [Miller 2005]. This
may be due to the lack of financial incentive. The Ontario government currently places a levy of only
six cents per tonne on virgin aggregate.
Another barrier relates to the quality of the RCA. Highways require quality material that meets
engineering, economic and environmental considerations [Turley 2003]. However, where high-
performance concrete is not required, RCA can be used and virgin aggregate conserved [Meyer
2008]. An excess amount of fine RCA is created during the crushing process. This excess fine
aggregate requires disposal or an alternate use. Depending on the source and type of RCA, the
absorption, strength, and impurities vary. This can mean that it is unusable or that it might adversely
2
impact the new pavement structure. Durability performance of RCA is not well understood because
of the limited and contradictory research results [Salem 2003]. Concrete that contains RCA has
decreased compressive strength and flexural strength, increased dry shrinkage, decreased sulfate
resistance and decreased chloride resistance. There is also a lack of knowledge on how RCA affects
durability since most studies focus only on the properties of RCA concrete [Olorunsogo 2002].
Government agencies have been slow to embrace the use of RCA due to concerns about quality
and a reluctance to change what has worked in the past [Turley 2003]. The use of material
specifications are a barrier to the use of RCA. A performance-based or end result specification would
allow more RCA use [Turley 2003]. However, specific standards on how to use RCA in new concrete
are not currently available.
1.2 Thesis Scope and Objectives
This research has the following objectives.

1. Highlight the current issues associated with using RCA.
2. Provide a detailed literature review which outlines sustainability, RCA issues, material
testing, mix design issues, concrete pavement design methods, data sources, and typical
pavement performance values.
3. Determine whether RCA can be used as a viable aggregate source for concrete pavements in
the Canadian environment.
4. Quantify the physical properties of four sources of RCA and assess their influence on
concrete pavement performance.
5. Compare the RCA properties and concrete performance to virgin aggregate using statistical
analysis.
6. Compare volumetric, property, performance, and functionality of different types of mix
designs using RCA.
7. Identify tests that correlate with initial and long term performance for quality control
purposes. Based on these tests, end result specifications and performance based
specifications will be developed.
8. Develop input values for RCA for the use in the new Mechanistic-Empirical Pavement
Design Guide (ME-PDG).
9. Detect pavement problems related to RCA use and provide guidelines on innovative tools and
technologies for corrective measures that can be taken.
10. Identify mix design properties for RCA and relate to long term pavement performance.
11. Demonstrate the use of RCA through the design and construction of test sections/slabs to
monitor RCA performance in the Canadian environment.
12. Develop a realistic life cycle cost model for RCA use in the pavement structure.
3
Figure 1.1 Research Framework
1.3 Project Significance
This research provides the provincial departments of transportation, state highway agencies, and the
concrete pavement industry with information on using RCA as a source of aggregate for new concrete
roadways and other applications in environments that have cold temperatures. Specifically, this
research addresses the technical and cost-effective concerns regarding RCA use in the Canadian
environment.
4
Chapter 2
Literature Review
2.1 Sustainability
Sustainability is defined by the World Commission on Environment and Development as Meeting

the needs of the present without compromising the ability of the future generations to meet their own
needs [Naik 2005]. One author describes achieving sustainability as the greatest challenge facing
the concrete industry in the 21st century [Mehta 2001]. He claims that the industry has a short-term
view point on the consumption of natural resources [Mehta 2001] and that in a finite world the
model of unlimited growth, unrestricted use of natural resources and uncontrolled pollution of the
environment is a recipe for planetary self-destruction. The Factor of Ten Club states that Within
one generation, nations can achieve a ten-fold increase in the efficiency with which they use energy,
natural resources and other materials [Mehta 2001].
There are three keys to sustainable development in the concrete industry [Mehta 1999]. First,
conserve concrete making material. This can be achieved by recycling aggregate by crushing
demolished concrete. Also, using recycled water from mixing plants and wash water from trucks
would decrease the need for fresh mixing water. Finally, using byproducts, such as fly ash, slag and
silica fume, from other industries reduces the amount of cement needed in the concrete.
Second, to aid in sustainable development, concrete structures need improved durability.
Sustainable concrete structures minimize the short and long-term societal impacts; however, to
achieve this durable concrete is needed [Naik 2005]. The current thinking is that designing for high
strength means durable concrete is achieved; however, designing concrete for durability and achieve
the necessary strength could also potentially improve sustainability. Concrete designs are needed that
minimize the greatest causes of deterioration such as corrosion, exposure to freeze/thaw, alkali-silica
reaction and sulfate attack [Mehta 1999]. Decreasing the permeability of the concrete through the use
of supplementary cementitious materials (SCMs) is an option.
Third, in order to achieve sustainable development training and education must be improved. A
1995 survey of Civil engineering departments showed that less than half of the responding schools
have an optional full semester course on concrete technology. To properly educate tomorrows
engineers in schools today, North American students need more education on cement and concrete
topics [Mehta 1999].
2.2 Recycled Concrete Aggregate
2.2.1 Producing Recycled Concrete Aggregate
Figure 2.1 shows a closed-loop concrete system [Kuroda 2005].
5
Figure 2.1 Closed-Loop Concrete System
Crushing concrete for use as RCA uses similar equipment and processes as when preparing virgin
aggregate. There are two types of crushers: compression and impact. Figure 2.2 shows both a cone
compression crusher and a jaw compression crusher [ACPA 2003]. Figure 2.3 shows a vertical and a
horizontal impact crusher where repeated blows against break plates reduce the size of the concrete
pieces [ACPA 2003].
Figure 2.2 Cone and Jaw Compression Crushers
Figure 2.3 Vertical and Horizontal Impact Crushers
6
Defects and irregular voids can be reduced by over 50% by sending RCA through a jaw or impact
crusher twice. Additional mechanical grinding will remove adherent mortar improving physical
properties while only introducing a negligible amount of new cracking. Cracking of the interface
transition zone was not affected significantly [Nagataki 2004].
2.2.2 Current Use of Recycled Concrete Aggregate
Many countries successfully use RCA including the United States, South Africa, Netherlands, United
Kingdom, Germany, France, Russia, Canada, and Japan [Olorunsogo 2002]. Currently, RCA is used
as an aggregate in granular subbases, lean-concrete subbases, soil-cement, and in new concrete as the
only source of aggregate or as a partial replacement of new aggregate [CAC 2004] [Kuo 2002,
Masood 01, ACPA 1993]. The Ministry of Land, Infrastructure, and Transportation has been
instrumental in Japan recycling 96% of the nations concrete waste through initiatives Recycling Plan
21 and Construction Recycling Promotion Plan 97 [Noguchi 2005]. Japan developed a special
technique that removes the original mortar from the concrete. This technique produces only 20
35% coarse aggregate compared to the 60 70% coarse aggregate that is produced in the current
system because of the large amount of adhered mortar [Dosho 2005].
In 2002, 28 states used RCA in pavement construction, 26 states use RCA as base or subbase
material only and two states allow for subbase use only [Kuo 2002]. By 2004, 41 of the 50 states are
recycling waste concrete into aggregate [FHWA 2004]. Figure 2.4 shows the 38 states where
recycled material is used as aggregate in road construction base material. Figure 2.5 shows the eleven
states where RCA is used in new concrete.
Figure 2.4 United States Using RCA as Base Material
Figure 2.5 United States Using RCA as Concrete Aggregate

7
The states leading the way in the use of RCA are Texas, Virginia, Michigan, Minnesota, and
California [FHWA 2004]. In 1980 the Minnesota Department of Transportation saved approximately
$600,000 by recycling sixteen miles of plain concrete pavement on US-59 [Salem 2003]. It is
estimated that using RCA saves approximately $4.80 per m2 ($4 y2) [ACPA 1993]. A US geological
survey conducted in 2000 showed that of the approximately 100 million tons of RCA produced
annually, 68% is used as a broad base, six percent is used in new concrete, nine percent is used in
asphalt, fourteen percent as riprap and other fill and seven percent in other uses [Li 2005].
In Canada there is very limited use of RCA. Ontarios non-virgin aggregate consumption is
estimated at three percent [Miller 2005]. In general, the Ontario Ministry of Transportation (MTO)
does not recycle old concrete into new concrete. Currently, the MTO is recycling old concrete by
blending it with virgin granular material for use as base and subbase [Gilbert, 2004]. The MTO
focuses on cost savings rather than on recycling. However, competitive bids in some areas of the
province where virgin aggregate is very expensive usually include recycling the existing pavement
[Gilbert, 2004].
2.2.3 Properties of RCA
Working with RCA can be challenging since often the specifics about the original concrete are
unknown [Oikonomou 2005]. Recycled concrete aggregate is highly heterogeneous and porous, with
a large amount of impurities. This makes it difficult to model and predict the resulting concrete
properties [Zaharieva 2003]. Better characterization of the properties of RCA would increase the
confidence needed to use RCA in new rigid pavements [Cuttell 2008].
2.2.3.1 Shape, Texture, and Gradation
In general, RCA has 100% crushed faces [Salem 2003]. The age and strength at which concrete is
crushed does not influence the amount of mortar attached to the aggregate or the gradation of the
RCA [Katz 2003]. Coarse RCA material contains about 6.5% adherent original mortar and the fine
material contains about 25% [Katz 2003].
2.2.3.2 Specific Gravity
Specific gravity or relative density is defined by the ASTM as the ratio of the density of a material to
the density of distilled water at a stated temperature. ASTM C 128 is the procedure for obtaining
specific gravity. Virgin aggregate has a specific gravity of 2.7 and RCA is 2.4. This difference is due
to the relative density of the old mortar attached to the RCA [Salem 2003] [Katz 2003]. Coarse RCA
typically has a specific gravity between 2.2 and 2.6 for saturated surface dry conditions. This value
decrease as the particle size decreases. Fine RCA has a specific gravity between 2.0 and 2.3 for
saturated surface dry conditions [Katz 2003, ACPA 1993].
8
2.2.3.3 Absorption
The ASTM defines absorption as the increase in mass of aggregate due to water penetration into the
pores of the particles during a prescribed period of time, but not including water adhering to the
outside surface of the particles, expressed as a percentage of the dry mass. Using the ASTM C 128
process virgin aggregate has a lower absorption of 0.3%. Coarse RCA has absorption of 2-6% and
fine RCA has an even higher absorption of 4-12% [Katz 2003, Kerkhoff 2001, ACPA 1993]. This
difference is due to the higher absorption of the old mortar contained in the RCA [Salem 2003]
[ACPA 1993].
2.2.3.4 Abrasion Resistance
Abrasion resistance is used as an index of aggregate quality and its ability to resist weathering and
loading action [CAC 2002]. Abrasion resistance of RCA is twelve percent lower than virgin
aggregate [Sagoe-Crentsil 2001]. The abrasion resistance results are not dependent on full or partial
RCA use [Abou-Zeid 2005]. Abrasion resistance for RCA ranges between 20-45% with an upper
range at 50% [ACPA 1993].
2.2.3.5 RCA Fines
Creation of high quality RCA produces a large amount of fines that can be problematic to deal with
[Naik 2005]. RCA fines can be mixed with a clay soil to improve the soil properties. Although the
addition of RCA fines did not significantly impact the clay soils liquid limit initially or after 21 days,
it almost doubled the plastic limit. The plasticity index at 21 days was 17.6 for clay soil containing
RCA fines compared to 35.9 for the control. This improves the soil classification from clay to silty
sand [Hansen 1986].
2.2.4 Properties of RCA Concrete
The cement mortar that is a part of the RCA significantly impacts the characteristics and performance
of the RCA containing concrete [Sagoe-Crentsil 2001]. Removal of some of the adherent mortar
helps to improve the properties of RCA containing concrete. The properties of the original concrete
have a significant influence on the properties of the RCA containing concrete (compressive strength,
tensile strength, bond stress at failure, F/T resistance) [Ajdukiewicz 2002].
There is a general lack of knowledge about how RCA use affects the durability of concrete. This is
due to contradictory research results and studies focused only on the properties of RCA containing
concrete not durability [Salem 2003]. RCA containing concrete performed in a comparable manner
to virgin concrete in terms of strength and durability [Shayan 2003, Olorunsogo 2002]. In general,
concrete durability is reduced as RCA content is increased [Olorunsogo 2002]. The increased
absorption of the RCA leads to larger amounts of shrinkage and cracking in RCA containing concrete
[Mesbah 1999]. However, durability properties can be improved with longer curing periods
[Olorunsogo 2002].
9
2.2.4.1 Workability
Concrete workability is defined as the effort required to manipulate a freshly mixed quantity of
concrete with minimum loss of homogeneity [Mehta 2006]. After five to ten minutes, RCA mixes are
stiffer and lose workability at a faster rate than mixes containing virgin aggregate [Salem 2003].
2.2.4.2 Slump
Slump is defined as the measure of the consistency of freshly mixed concrete, equal to the
immediate subsidence of a specimen molded with a standard slump cone [CAC 2002]. Admixtures
in the RCA had no significant impact on the slump of the new RCA concrete [Hansen 1984]. The
more RCA that is used in a cement mix, the higher the w/c ratio that is needed. This will result in a
higher slump [Lin 2004]. However, assuming a constant w/c ratio, RCA concrete mixes have a
decrease in slump compared to virgin concrete mixes. RCA has a higher absorption and an angular
texture that increases the internal friction [Rashwan 1997]. As the amount of RCA increases at a
constant w/c ratio, the workability decreases [Topcu 2003]. The moisture state of the RCA impacts
the slump and slump loss of the concrete. Keeping a constant w/c ratio, slump and slump loss was
the highest for concrete that contained oven-dried RCA as compared to air-dried or saturated surface
dry RCA.
2.2.4.3 Air Content
Air content of freshly mixed concrete is based on a change of volume for a change in pressure. RCA
has a higher void content than virgin aggregate. This results in RCA containing concrete having a
larger amount of entrapped air compared to virgin aggregate [Salem 2003]. The air content of RCA
containing concrete is higher than the concrete from which the RCA was made since the new concrete
contains both the air entrapped in the concrete and the air in the RCA [Katz 2003]. Admixtures in the
original concrete that is made into RCA had no significant impact on the air content of the new RCA
concrete [Hansen 1984].
2.2.4.4 Initial Set Time
The time required for the cement paste to ceases being fluid and plastic is the initial set time [CAC
2002]. Admixtures in the original concrete had no significant impact on the initial set time of the new
RCA concrete [Hansen 1984].
2.2.4.5 Final Set Time
The final set time is the time required for the cement paste to develop a certain degree of strength
[CAC 2002]. There was no significant difference in final set time for RCA containing concrete when
the RCA was made from a concrete containing an admixture [Hansen 1984].
10
2.2.4.6 Compressive Strength
Compressive strength is the ability to resist compression loads [CAC 2002]. In general, using RCA
in the concrete mix decreases compressive strength compared to virgin aggregate. However, at 28
days, all mix designs usually exceed 50 MPa compressive strength [Shayan 2003]. One study
showed the compressive strength of virgin concrete was 58.6 MPa, and the RCA concrete ranged
from 50.9 to 62.1 MPa. There were higher values for concrete made with 50% RCA compared to
100% RCA [Poon 2002]. The loss of compressive strength is in the range of 30-40% for the concrete
made with RCA at 28-days [Katz 2003]. There was a minor reduction in 28- and 56-day compressive
strength when virgin aggregate was partially replaced with RCA and a much greater reduction when
RCA was used in full [Abou-Zeid 2005].
The most influential parameter affecting compressive strength is the w/c ratio [Lin 2004]. Other
influential parameters include fine RCA content, cleanness of aggregate, interaction between fine
RCA content and crushed brick content, and interaction between w/c ratio and coarse RCA content
[Lin 2004]. Keeping a constant w/c ratio, air-dried RCA containing concrete had the highest
compressive strength compared to oven-dried and saturated surface dry RCA [Poon 2003]. Using
unwashed RCA reduces compressive strength particularly at lower w/c ratios. Compressive strength
is 60% of virgin concrete at 0.38 w/c and 75% at 0.6 w/c [Chen 2002].
There seems to be a strong interaction between maximum aggregate size and water-cement ratio
when compared with compressive strength development [Tavakoli 1996a]. Compressive strength
may increase for RCA due to a lower w/c ratio compared to virgin aggregate, 14% and 34%
respectively. However, compressive strength may decrease for RCA since it has a higher air
entrainment, 25%, compared to virgin aggregate 23% [Salem 2003].
The majority of strength loss for RCA concrete can be attributed to material smaller than 2 mm
because natural sand has greater strength than RCA fines [ACPA 1993]. It is recommended to keep
RCA fines less than 50% of the sand content [Shayan 2003]. Bonding between the RCA and the
cement can be affected by loose particles created during the crushing process. Treating the RCA by
impregnation of silica fume resulted in an increase in compressive strength of approximately at 30%
at 7-days and 15% at 28-days. Exposing the RCA to ultrasound resulted in a uniform increase of 7%
compressive strength over time [Katz 2004]. The age at which the RCA is crushed has a significant
impact on the compressive strength of the final concrete. For example, crushing concrete into RCA
after three days compared to one day resulted in a seven percent increase in compressive strength of
the new RCA concrete at 7 days. The difference in compressive strength of the new RCA concrete
increased to 13% when measured at 90 days [Katz 2003]. The compressive strength of the original
crushed concrete influences the compressive strength of the RCA concrete [Tavakoli 1996a].
However, it has been reported that RCA concrete can produce higher compressive strengths than the
original concrete [Ajdukiewicz 2002]. For example, an 80+ MPa concrete was created from an
original 60MPa concrete [Ajdukiewicz 2002].
When comparing laboratory-made RCA and field demolished RCA, there was the same basic trend
in all strength development [Tavakoli 1996a]. Admixtures in the original concrete had no significant
impact on the compressive strength of the new RCA concrete [Hansen 1984]. When slag is added to
the RCA concrete, it develops strength over a longer period of time compared to normal concrete
11
[Sagoe-Crentsil 2001]. Some research suggests that compressive strength is dependant on the amount
of time the RCA spent in the stockpile after crushing [Rashwan 1997]. For example, concrete made
with RCA that was in the stockpile one day had a 25% higher compressive strength than concrete
made with RCA that was in the stockpile 28 days. Concrete made with RCA that was in the stockpile
seven days had a seven percent lower compressive strength than concrete that was in the stockpile 28
days [Rashwan 1997]. RCA concrete showed good performance when exposed to temperatures up to
600oC with a loss in compressive strength of 20-25% [Abou-Zeid 2005]. When RCA concrete fails it
is usually because cracks passed through the RCA: however, when virgin concrete fails it is usually
due to bond failure at the aggregate-paste interface [Salem 2003].
2.2.4.7 Flexural Strength
Flexural strength or modulus of rupture is the ability to resist tension resulting from bending [CAC
2002]. There are conflicting results about how RCA use affects flexural strength. The results range
from RCA decreasing flexural strength [Zaharieva 2004, Katz 2003, Salem 2003] to RCA increasing
flexural strength [Poon 2002]. One study showed a decrease in flexural strength between 10-20%
[Zaharieva 2004]. Other studies found comparable flexural strength results between RCA concrete
and the control [Tavakoli 1996a, Abou-Zeid 2005]. And yet another study showed that flexural
strength increased with the amount of RCA used. Virgin concrete had a flexural strength of 3.31
MPa, and RCA concrete ranged from 3.74 to 3.89 MPa with 100% RCA concrete having higher
values than 50% RCA concrete [Poon 2002].
The parameters that influence flexural strength are not completely clear. However, minor
decreases in strength can be attributed to material smaller than 2 mm resulting from natural sand
having greater strength than RCA fines [ACPA 1993]. One study suggested that flexural strength
was comparable to the w/c ratio [Tavakoli 1996a].
Flexural strength can be predicted by using ACI 363R equation.
f r' = 0.94 f c' (2.1)
where f r' = flexural strength (MPa)

f c' = compressive strength (MPa)
!
However, the actual flexural strength is higher than what is predicted by ACI 363R equations [Katz
! 2003, Tavakoli 1996a] and the variation increases as the w/c ratio is increased [Tavakoli 1996a].
!
2.2.4.8 Bond Strength
Bond strength is defined as the force required to break two materials apart [CAC 2002]. On average
RCA concretes failed with a 20% lower force than virgin concretes [Ajdukiewicz 2002].
12
2.2.4.9 Interfacial Transition Zone
The interfacial transition zone (ITZ) is a small region next to the particles of coarse aggregate. This is
considered the strength-limiting phase in concrete [Mehta 2006]. The ITZ is an important factor in
compressive strength development in RCA concrete [Poon 2004]. RCA cement has a higher ITZ due
to higher porosity and absorption of the RCA resulting in a lower compressive strength [Poon 2004].
Normal RCA concrete had a loose and porous ITZ of 30-60 m. Where as RCA concrete made from
high-performance concrete had a dense ITZ of 10 m [Poon 2004]. The ITZ strength, Vickers
hardness, is related to the w/c ratio. RCA concrete with a high w/c ratio (0.55 and 0.7) produces a
weaker ITZ than the original concrete from which the RCA is made; however, a low w/c ratio (0.25
and 0.4) produces an ITZ that is greater than the original concrete strength [Otuki 2003].
2.2.4.10 Hardness
The hardness of concrete is tested by checking at the surface for concrete uniformity. This correlates
to compressive strength and stiffness [CAC 2002]. The Schmidt hardness concrete values decreased
from 21.3 MPa for virgin aggregate to 11.6 MPa for 100% RCA concrete [Topcu 1997]. This decline
in hardness usually corresponds to a decrease in compressive strength [Topcu 1997].
2.2.4.11 Modulus of Elasticity
The ratio of normal stress to corresponding strain is called the modulus of elasticity [CAC 2002].
Use of RCA causes a decrease in the modulus of elasticity [Katz 2003, Salem 2003, Salem 1998,
Hansen 1985]. Research shows that a 20-30% decrease in modulus of elasticity is common when
using RCA but up to a 40% decrease can be expected [Chen 2002, Hansen 1985]. Washing the
aggregate did not cause any significant variation in modulus of elasticity values [Chen 2002].
Another study showed that even with the addition of fly ash and an increase in air content, the
modulus of elasticity of the RCA concrete was lower compared to the virgin concrete [Salem 1998].
The actual modulus of elasticity results are approximately 25% lower than those calculated by the
ACI Equation [Katz 2003].
E c = 0.043W c1.5 f c' (2.2)

where E c = modulus of elasticity
W c = correction factor for the lower density of RCA
!
f c' = compressive strength (MPa)
!
! 2.2.4.12 Freeze-Thaw Resistance
! Freeze-thaw resistance (F/T) is the ability of concrete to withstand cycles of freezing and thawing and
is one of the measures of durability [CAC 2002]. Freeze-thaw resistance of RCA has produced mixed
results [Gokce 2004, Salem 1998].
13
Increasing air content is the single most effective method of improving F/T resistance [Salem 1998]
[Salem 2003]. It is difficult to produce a F/T resistant concrete from RCA containing no air-
entrainment, therefore air-entrainment is recommended [Gokce 2004]. Air-entrained concretes have
lower amounts of mass loss than non air-entrained concretes [Gokce 2004]. RCA made from air-
entrained concrete performed better than the original concrete aggregate having a relative dynamic
modulus of elasticity above 90% at 500 cycles. Where as RCA made from non air-entrained concrete
performed poorly, the relative dynamic modulus of elasticity dropped below 60% at 90 F/T cycles
[Gokce 2004]. Microscopic examination showed that the interfacial transition zone (ITZ) cracking
ratio was significantly higher for all non air-entrained concretes [Gokce 2004]. Micro-crack
development and propagation was due to the non air-entrained adherent mortar [Gokce 2004].
RCA concrete had a greater weight gain compared to virgin concrete. This is due to having a
higher level of absorption and permeability [Salem 2003]. Use of SCMs can improve the F/T
resistance of the RCA since it decreases permeability [Gokce 2004]. Increasing fly ash content
significantly improved F/T resistance [Salem 1998]. Adding other SCMs, such as silica fume and
10% metakolin, to non air-entrained concrete improved the F/T resistance. Only the metakolin was
able to improve the durability enough to exceed 300 cycles [Gokce 2004].
There are several other factors that have been identified as affecting F/T resistance. First, reducing
the amount of adherent mortar on the RCA results in a limited benefit [Gokce 2004]. Second, RCA
made from high performancy concrete resulted in improved F/T resistance for the RCA concrete
[Ajdukiewicz 2002]. Third, decreasing the w/c ratio also improves F/T results [Salem 2003]. Finally,
presoaking the RCA had a negative effect on F/T resistance of the RCA concrete. The best F/T
resistance was in concrete made with only coarse RCA and virgin fines, followed by not presoaked,
full RCA concrete and last full RCA concrete that was presoaked [Zaharieva 2004].
2.2.4.13 Dry Shrinkage
The loss of moisture from the surface of hardened concrete leading to slab shrinkage may result in
cracking [Mehta 2006]. RCA concrete has a greater amount of dry shrinkage compared to virgin
concrete [Ajdukiewicz 2002, Tavakoli 1996b, Hansen 1985]. The increase in dry shrinkage for RCA
concrete compared to virgin concrete ranges from 20-90% [Hansen 1985]. One study showed that
RCA concrete had a dry shrinkage of 0.7 to 0.8 mm/m compared to virgin concrete with a dry
shrinkage of 0.27 mm/m [Katz 2003]. The amount of dry shrinkage increases with the amount of
RCA used in the concrete [Poon 2002]. The method of dry shrinkage is the same for RCA and virgin
aggregate [Tavakoli 1996b].
Metallic fibres can be used to decrease strain at the interface between dry and moist concrete
thereby reducing the number and severity of dry shrinkage cracks. At 600 days, RCA concrete
containing metallic fibres had a 15% reduction in dry shrinkage compared to RCA concrete without
metallic fibres [Mesbah 1999]. Polypropylene fibres actually increased dry shrinkage [Mesbah
1999]. When tested for restrained shrinkage, metallic fibres were most effective in minimizing the
number of cracks and crack width compared to polypropylene fibres or no fibres. After 50 days,
RCA concrete with 1% metallic fibres had a crack width of 0.14 mm, RCA concrete with 15%
14
polypropylene had a crack width of 0.24 mm and RCA concrete without metallic fibres had a crack
width of 1.07 mm [Mesbah 1999].
There are several other factors that affect dry shrinkage. First, the amount of dry shrinkage was
increased when 20% slag was added to the RCA concrete using the same w/c ratio [Sagoe-Crentsil
2001]. Second, increasing w/c ratio causes an increase in shrinkage [Tavakoli 1996b]. Third, the
more mortar attached to the RCA the greater the amount of shrinkage. This is due to the increase in
water absorption [Tavakoli 1996b]. Finally, increasing the maximum size of the RCA reduces the
amount of shrinkage [Tavakoli 1996b]. The exact properties of the RCA are dependant on the
original concrete and the RCA [Tavakoli 1996b].
2.2.4.14 Alkali-Aggregate Reactivity
Aggregates that contain certain forms of silica or carbonates react with alkali hydroxides in the
concrete producing an expansive gel [CAC 2002]. RCA concrete had increased AAR susceptibility.
However, at one year the RCA mixes are considered non reactive [Shayan 2003]. Adding 25% fly
ash to RCA concrete mitigated the alkali-silica reactivity to acceptable levels; however, fifteen
percent fly ash was not significantly improve levels [Li 2006].
2.2.4.15 Corrosion
The corrosion potential of steel is found using the half cell potential test which measures the voltage
difference between the steel and a reference electron [Mehta 2006]. Corrosion usually results from
exposure to sulfate, chloride and carbonates. Both virgin concrete and RCA concrete had a very low
corrosion risk using the half cell potential test [Shayan 2003].
2.2.4.16 Depth of Carbonation
Carbonation is the process by which carbon dioxide in the air penetrates the concrete and reacts with
hydroxide to form carbonates. Once carbonates form, they decrease the pH of the concrete resulting
in decreased corrosion protection for the reinforced steal [CAC 2002]. The depth of carbonation is
1.3-2.5 times greater for RCA concretes compared to virgin concretes [Katz 2003, Shayan 2003].
2.2.4.17 Sulfate Resistance
The ability of concrete to resist penetration of sulfates from soil or water that reacts with the hardened
cement paste resulting in strength loss [CAC 2002]. The standard for sulfate resistance, according to
ASTM C 157, is expansion should be less than 0.1% at six months, or for high resistance expansion
should be less than 0.05% at 6 months and 0.1% at one year [Shayan 2003]. RCA has better sulfate
resistance than virgin aggregate. Generally expansion is below 0.025% at one year [Shayan 2003].
15
2.2.4.18 Chloride Content
The ACI 201 recommends that cement for reinforced concrete in a moist environment and exposed to
chloride have less than 0.1% soluble chloride ion, cement for reinforced concrete in a moist
environment without exposure to chloride have a less than 0.15% soluble chloride ion, and concrete
in a dry environment has no recommended limit [Hansen 1984]. Chlorides are usually introduced
from concrete that was exposed to deicing chemicals. Coarse RCA had a chloride content of 0.07 to
0.09% and fine RCA had an approximate chloride content of 0.03% [ACPA 1993]. Both of these
values are below the ACI recommended values. However, one study reported 0.69% soluble chloride
ion by weight of cement for RCA concrete which is above the ACI recommended value [Hansen
1984].
2.2.4.19 Chloride Penetration
Chloride penetration is the depth to which chloride penetrates when concrete is exposed to a chloride
source such as deicing chemicals. This results in corrosion of the steel reinforcement. Penetration for
RCA concrete was 2.3 mm and 2.2 mm deeper than virgin concrete for a fifteen percent NaCl
solution for ten cycles and a five percent NaCl solution continuous for 112 days respectively [Shayan
2003].
2.2.4.20 Chloride Conductivity
Chloride conductivity is the rate at which chloride ions diffuse into the concrete [Olorunsogo 2002].
Conductivity increases as RCA content increases [Olorunsogo 2002]. After curing for 56 days, virgin
concrete and 50% RCA concrete had a very poor chloride conductivity rating and 100% RCA had a
good rating [Olorunsogo 2002]. This is likely due to cracks and fissures in the attached mortar of
RCA allowing fluid to pass easily through [Olorunsogo 2002]. In general, conductivity decreases as
the length of curing increases [Olorunsogo 2002].
2.2.4.21 Water Absorption
Water absorption measures how much water is absorbed into the concrete by capillary action
[Olorunsogo 2002] [CAC 2002]. RCA concrete has higher water absorption. One study showed the
water absorption of RCA concrete to be 7.2% compared to 3.8% in the virgin concrete [Katz 2003].
A second study states that the water absorption of RCA concrete was, on average, 25% greater than
virgin concrete [Sagoe-Crentsil 2001]. As the amount of RCA in the concrete increases, so does the
water absorption [Olorunsogo 2002]. The longer the concrete is allowed to cure, the lower the water
absorption [Olorunsogo 2002].
2.2.4.22 Permeability
The permeability of concrete is a measure of the passage of fluids or gases [CAC 2002]. The
permeability of RCA concrete is highly variable and is approximately 10% to 45% higher than virgin
16
aggregate [Abou-Zeid 2005, Zaharieva 2003]. In general, RCA concrete has a higher w/c ratio due to
water absorption by the aggregate. This water evaporates during curing resulting in greater porosity
[Zaharieva 2003]. Increasing the length of curing creates a finer porosity reducing permeability by
50% [Zaharieva 2003].
2.2.4.23 Skid Resistance
Skid resistance is a measure of the frictional characteristics of the pavements surface [TAC 1997].
Virgin concrete has a skid resistance of 98 British Pendulum Number (BPN) where as RCA concrete
is higher, ranging from 108 BPN to 114 BPN. There was no significant difference in the skid
resistance values when different amounts of RCA where used or when fly ash was added [Poon
2002].
2.2.4.24 Ultra Sound Pulse Velocity
Ultrasound pulse velocity of virgin concrete is approximately 69-70 s and increases to 92-93 s for
RCA concrete due to a widening of the air voids and a decrease in strength [Topcu 1997].
2.2.4.25 Coefficient of Thermal Expansion
There is conflicting research on the impact of RCA on the coefficient of thermal expansion (CTE).
One study showed that RCA containing concrete had higher CTE values of 8.9x10-6 /oC for a cylinder
and 11.6x10-6 /oC for a prism compared to other types of coarse aggregate [Yang 2003]. Where as a
Federal Highway Administration (FHWA) review of concrete containing RCA identified CTE as one
of the characteristics that improved compared to concrete using virgin aggregate [FHWA 2004,
Turley 2003].
2.2.4.26 Pore Structure
The average pore radius increases as RCA content increases and is most significant at early age (7-
days) and decreases over time (28- and 90-days). Reduction over time is due to the continued
hydration of the cement. For example, the pore radius at 90 days for virgin concrete is 18.8 nm, for
30% RCA concrete is 19.6 nm, for 60% RCA is 21 nm, and for 100% RCA is 24.7 nm [Gomez-
Soberon 2002].
2.2.5 RCA Properties Summary
A summary of RCA properties is provided in Table 2.1. Table 2.2 summarizes the properties of RCA
concrete.
17
Table 2.1 RCA Material Properties Summary
Aggregate Material Properties Recycle Concrete Aggregate Key Point
Shape and texture 100% crushed faces
6.5% adherent mortar (coarse) Amount of adherent mortar will
affect concrete properties
25% adherent mortar (fine)
Specific gravity Decreases up to 25% (coarse)
Decreases up to 35% (fine)
Absorption 2.0 6.0% (coarse)
4.0 12.0% (fine)
Abrasion resistance Decreases up to 12%
Table 2.2 RCA Concrete Properties Summary

Concrete Material Properties Recycle Concrete Aggregate Key Point
Workability Decreases at a faster rate
Slump Decreases at same w/c ratio Larger water demand for RCA
Air content Increases total air content Larger volume of entrapped air
Initial set time Comparable results
Final set time Comparable results
Compressive strength Decreases up to 40% Results based on maximum
aggregate size and w/c ratio
Flexural strength Comparable to decreases up to Influential parameter is not
20% clear
Bond strength Decreases up to 20%
Interfacial transition zone More porous and less dense Higher porosity and absorption
Hardness Decreases up to 45% Corresponds to a decrease in
compressive strength
Modulus of elasticity Decreases up to 40%
Freeze-thaw resistance Variable results Improvements by decreasing
w/c ratio and adding a SCM
18
Concrete Material Properties Recycle Concrete Aggregate Key Point
Dry shrinkage Increases up to 90% Increases with the amount of
RCA used
Alkali-aggregate reactivity Increased susceptibility
Corrosion Comparable results Based on half-cell potential
reading
Depth of carbonation Increased depth 1.3 2.0 times Lower pH levels
Sulfate resistance Improves resistance
Chloride content Comparable results
Chloride penetration Increases depth 2.2 2.3 mm
Chloride conductivity Increases Increases with the amount of
RCA used
Water absorption Increases up to 25% Increases with the amount of
RCA used
Permeability Increases up to 45% Higher w/c ratio and absorption
Skid resistance Comparable results
2.3 RCA Concrete Mix Designs
The main concern with using RCA in concrete mixes is the long-term properties of the finished
product and the cost of using RCA [Abou-Zeid 2005].
2.3.1 Components of concrete mixes
2.3.1.1 Aggregate
Many differences between RCA and virgin aggregate have been discussed in the previous section;
however some studies show the differences to be minimal. One study showed no significant
difference between the RCA concrete and the virgin concrete at 28-days [Sagoe-Crentsil 2001].
Another study showed that the addition of 20% coarse RCA had no significant difference in the
properties and performance of the new concrete [Oikonomou 2005]. The main reason for differences
in RCA is due to the attached mortar. There is a relationship between an increase in mortar resulting
in a decrease in performance [Tavakoli 1996a]. Another reason for performance differences of RCA
concrete is due to the size of the aggregate. Too many RCA fines results in an unworkable mix. The
properties of RCA concrete are improved when natural sands are used [Ajdukiewicz 2002]. There are
several recommendations in the literature regarding the amount of RCA that should be used in
19
concrete mixes. These recommendations include using a maximum of 30% coarse RCA to meet
strength criteria [Topcu 2003] and a maximum of 50% coarse RCA [poon 2003]. Also, it is
suggested that RCA fines be limited to 10 to 20% and natural sand be used for the rest [ACPA 1993].
The RCA fines do not qualify as hydraulic cement [Hansen 1983]. Reducing the maximum size of
the aggregate to nineteen millimeter can decrease D cracking [ACPA 1993].
2.3.1.2 SCM
There is conflicting evidence as to how SCMs impact RCA concrete. One study reports that the
addition of fly ash reduced compressive strength, flexural strengths and skid resistance while
improving dry shrinkage [Poon 2002]. The ACPA suggests that the addition of 20% fly ash can
enhance durability [ACPA 1993]. Silica fume greatly improves the concrete strength [Hansen 1983].
2.3.2 Mixing Methods
There are two mixing methods used in making concrete, conventional and double mixing method.
The double mixing method divides the water into two equal proportions that are added at different
stages of the mixing process. This creates areas of high and low w/c ratio paste thereby improving
the characteristics of the concrete. The double mixing method increases compressive strength, tensile
strength and ITZ strength and decreases ITZ thickness, depth of chloride penetration and depth of
carbonation. [Otsuki 2003, Ryu 2002].
2.3.3 Testing
2.3.3.1 Aggregate Testing
Table 2.3 presents two existing RCA specifications for use in concrete, the Greek Specification of
Concrete Technology (GSCT) and the Japanese Industrial Standards (JIS), and one proposed
specification limits for Egypt.
Since RCA is of similar standards to virgin aggregate, it should also be tested for: grain size,
specific gravity, density, water absorption, and abrasion resistance [Topcu 1995]. The GSCT also
draws attention to the importance of investigate the presence of chlorides, to reduce the potential for
an alkali-silica reaction but does not give any specific limits.
20
Table 2.3 Summary of RCA Specifications
GSCT1 JIS2,3 Egypt4
Coarse Fine
Specific gravity (kg/m3) > 2.2 > 2.5 > 2.5

Water absorption (%) < 3.0 < 3.0 < 3.5 < 7.0
Foreign ingredients (%) < 1.0 < 1.0 < 1.0
Foreign ingredients (kg/m3) 2 10
Organic ingredients (%) < 0.5
Sulphate ingredients (%) < 1.0
Amount of sand (%) < 5.0
Amount of filler (%) < 2.0
LA abrasion (%) < 40.0 < 35.0 40 50
Soft granulars < 3.0
Soundness loss (%) < 10.0
Sand equivalent (%) > 80.0
Solid volume (%) < 55.0 < 53.0
Material passing 75 m (%) < 1.0 < 7.0
10% fineness value (kN) 50 150
Chloride content < 0.04 < 0.4
ASR Harmless Harmless
Flakiness index (%) 40
1. [Oikonomou 2005], 2. [Kuroda 2005], 3. [Noguchi 2005], 4. [Kamel 2008]
2.3.3.2 Concrete Testing
Table 2.4 shows the maximum amount of harmful substances that should be allowed in RCA concrete
[Oikonomou 2005].
Table 2.4 Maximum Limit of Harmful Substances Allowed in RCA Concrete

Element As Pb Cd Cr Cu Ni I Zn
Limit (mg/L) 50 100 5 100 200 100 2 400
21
2.3.4 Mix Designs with RCA
Standard mix designs have been created as shown in Table 2.5 [Kuroda 2005]. Based on the type of
RCA being used and the strength required, the w/c ratio and component amounts are given.
Table 2.5 RCA Mix Designs

Aggregate Type Strength w/c Unit Weight (kg/m3)
Coarse Fine (MPa) ratio Cement Water Coarse Fine
Recycled Recycled 27 0.517 323 167 1011 792
30 0.484 347 168 1011 769
33 0.455 374 170 1011 743
Virgin Recycled 27 0.544 320 174 972 829
30 0.504 345 174 972 808
33 0.470 372 175 972 783
Recycled Virgin 27 0.517 323 167 1080 722
30 0.484 347 168 1080 699
33 0.455 374 170 1080 673
Using the standard ACI mix design procedure with 100% RCA, the compressive strength at 28-
days was reduced by ten percent. Other mix design methods generally have a 20% to 30% reduction
in compressive strength [Bairagi 1990]. To compensate for the reduction in compressive strength
when using RCA, an additional eight to thirteen percent of cement can be added [Bairagi 1990].
Traditionally the cement content of RCA concrete is increased using conventional volumetric mix
proportioning ignoring the adherent mortar content of the RCA [Abbas 2008, Fathifaz 2008]. A new
approach, equivalent mortar volume (EMV), looks at RCA as a two phase material of original
aggregate and adherent mortar. This process accounts for a change in the mortar content, decreasing
the amount of cement required [Abbas 2008]. Equation 2.3 is used to determine the EMV of RCA
[Fathifazl 2008].
Using the EMV method, a conventional mix that requires 400kg/m3 of cement would be reduced to
337-354 kg/m3 of cement [Abbas 2008]. At a constant w/c ratio, the EMV method provides
improved fresh properties (slump and density), improved hardened properties (strength, elastic
modulus, and creep) and better durability (F/T, chloride penetration) compared to the conventional
method [Abbas 2008, Fathifazl 2008].
22
NAC
RAC VNA " (1# R)
VRCA = (2.3)
$ SGbRCA '
(1# RMC)& VA )
% SGb (
RAC
where: VRCA = volume fraction of the coarse RCA in recycled aggregate concrete
NAC
VNA = volume fraction of the natural aggregate in the normal concrete
!
R = is the volumetric ratio of natural aggregate in RAC to normal concrete
!
SGbRCA = bulk specific gravity of the RCA
!
SGbVA = bulk specific gravity of the virgin aggregate
!
!
! 2.3.5 Pavement Performance
There are several studies showing the performance of RCA containing concrete. A study of 5 state
highway agencies using RCA in paving applications showed comparable performance to conventional
pavement; however, a difference in load transfer efficiency was observed due to the adherent mortar
causing an increase in dry shrinkage and decrease in CTE [Cuttell 2008] . At the University of
Central Floridas Circular Accelerated Test Track no distresses were observed for the RCA mixes
(0%, 25%, 75%, and 100%) despite having lower compressive, flexural, and tensile strengths [Chini
2001]. The FHWA states that when appropriately used recycled materials can effectively and safely
reduce cost, save time, offer equal or, in some cases, significant improvements to performance
qualities, and provide long-term environmental benefits [FHWA 2007].
2.4 Implementation of RCA Use Steps in the Right Direction Going Green
What started as a demand by a few environmentalists has now been adopted by the public at large
and by governments on local, State, and national levels. Even owners and developers have discovered
that going green is not only accepted now as politically correct or as a source of intangible benefits
and good publicity, but it also is a way for them to improve their bottom line. [Meyer 2008]
2.4.1 FHWA Recycled Materials Policy
The FHWA has developed a Recycled Materials Policy as an initial step to increasing RCA use. The
Recycled Materials Policy is made up of the following five points [FHWA 2007].
1. Recycling and reuse can offer engineering, economic and environmental benefits.
2. Recycled materials should get first consideration in materials selection.
3. Determination of the use of recycled materials should include an initial review of engineering
and environmental suitability.
4. An assessment of economic benefits should follow in the selection process.
23
5. Restrictions that prohibit the use of recycled materials without technical basis should be
removed from specifications.
2.4.2 LEED Program
The Leadership in Energy and Environmental Design (LEED) program is a point rating system (69
maximum) to evaluate the environmental performance of a structure and encourage market
transformation towards sustainable design [USGBC 2005]. There are four levels of LEED
certification. Achieving 26-32 points is classed as certified, 33-38 points achieves silver, 39-51
points achieves gold, and over 52 points achieves the platinum level. The LEED rating system has
six categories: sustainable sites, water efficiency, energy and atmosphere, material and resources,
indoor environmental quality, and innovation and design process. By using concrete as a building
material it is possible to earn between 11-21 points. [PCA An engineers guide to: building green
with concrete]. Having RCA as an aggregate source contributes up to five points with the following
credits:
1. MR Credit 4.1 Recycled Content, 10% - 1 point
2. MR Credit 4.2 Recycled Content, 20% - 1 point
3. MR Credit 5.1 Regional Materials, 10% Extracted, Processed & Manufactured 1 point
4. MR Credit 5.2 Regional Materials, 20% Extracted, Processed & Manufactured 1 point
5. MR Credit 6 Rapidly Renewable Materials 1 point
If the RCA is obtained from an onsite source then an additional two points can be achieved due to
diverting construction waste from landfill sites. These credits are:
1. MR Credit 2.1 Construction Waste Management, Divert 50% from Disposal, and
2. MR Credit 2.2 Construction Waste Management, Divert 75% from Disposal.
2.4.3 Success Stories
2.4.3.1 Lester B. Pearson International Airport (LBPIA)
LBPIA is its own mini-ecosystem encompassing forests, grasslands, creeks, and streams, as well as
runways, terminal buildings, and roadways. The GTAA recognizes the need to preserve and protect
these natural resources while continuing to meet the demands of a growing air travel industry [GTAA
2007]. LBPIA was the first North American airport to achieve ISO 14001 certification in 1999. In
order to obtain this prestigious certificate an Environmental Management System was established for
all aspects of airport operations to monitor and improve environmental performance. To reach and
maintain this standing the GTAA set goals to make continual improvement, prevent pollution, and
comply with legislation. [Seow 2005] During demolition of Terminal 1 and more recently Terminal
2, targets called for a minimum of 80% and 85% respectively of C&D waste (concrete, rubble,
metals, wood, and other material) to be reused and recycled. Demolition of Terminal 1 produced
253,000 tonnes of concrete that was 100% recycled. [GTAA 2007, Seow 2005] Recycling was
24
achieved by establishing an on-site mobile crushing plant to produce 2 minus RCA backfill material
(similar to the gradation of a Granular B) for the new terminals apron subbase [Seow 2005].
2.4.3.2 Stapleton International Airport
In 1995, when Stapleton International Airport in Denver was decommissioned, enough concrete and
asphalt recycled material was produced to construct approximately 10,000 miles of two lane roadway
[Meyer 2008].
2.4.3.3 Texas Department of Transportation (TxDOT)
The TxDOT has used RCA as an aggregate source for new concrete pavements and as a base material
since the early 1990s. [Harrington 2004] The switch to using RCA was achieved through training and
information sessions to dispel the perception that RCA is a substandard or waste material.
2.4.3.4 Minnesota Department of transportation (MnDOT)
The MnDOT recycles close to 100% of the concrete from its pavements and uses it as a base material
[Harrington 2004].
2.4.3.5 Michigan Department of Transportation (MDOT)
The MDOT has successfully used RCA as a base and subbase material on I-75, I-94, and I-95, US 41
[Harrington 2004].
25
Chapter 3
Materials and Test Mixes
This chapter discusses the materials that are used in the research.
3.1 Aggregate Sources
3.1.1 Sources of Virgin Coarse Aggregate
In this research, two virgin coarse limestone aggregates were supplied by Dufferin Construction. The
37.5 mm minus aggregate came from Dufferins Acton Quarry, Sand, and Gravel Pit, and the 19.0
mm minus aggregate came from Dufferins Milton Quarry, Sand, and Gravel Pit.
3.1.2 Sources of Recycled Coarse Aggregate
The coarse recycled concrete aggregate (RCA) in this research was sourced from decommissioned
concrete sidewalks, curbs and gutters from the Region of Waterloo. The original curb and gutter
concrete had been cured properly with a maximum aggregate size of 19 mm, adequate air-
entrainment, and a compressive strength in excess of 30 MPa at 28-days. The concrete sidewalks,
curbs and gutters were chosen as the source or RCA since it did not contain foreign material and had
a maximum aggregate size of 19 mm. Since the concrete that the RCA was crushed from contained a
maximum aggregate size of 19 mm, this was the maximum size of the RCA that was used. Any RCA
over 19 mm would contain excess interfacial transition zones that would negatively impact the
strength of the concrete. Figure 3.1 shows the waste concrete prior to being crushed and the resulting
coarse aggregate that was used in the study.
Figure 3.1 Waste Concrete and Recycled Aggregate
26
3.1.3 Source of Virgin Fine Aggregate
Natural sand was used as fine aggregate since it provides a greater strength than fine RCA [ACPA
1993]. The fine RCA contains many impurities and results in strength loss in the concrete. Virgin
fine aggregate (natural sand) was supplied by Dufferin Construction from the Blair Quarry, Sand, and
Gravel Pit.
3.2 Aggregate Testing
3.2.1 MTO Guidelines for Aggregate
In Ontario, the material specification for aggregates to be used in concrete is given in MTO Ontario
Provincial Standards Specification (OPSS) 1002. The MTO gradation limits and properties for coarse
aggregates used in concrete pavements are given in Table 3.1 and Table 3.2 respectively. Table 3.3
and Table 3.4 summarize the gradation limits and properties for fine aggregate.
Table 3.1 Coarse Aggregate Gradation Limits

Sieve Size Percent Passing
37.5 mm 19.0 mm Combined
53.0 mm 100 - 100
37.5 mm 90-100 - 95-100
26.5 mm 20-55 100 -
19.0 mm 0-15 85-100 35-70
9.5 mm 0-5 20-55 10-30
4.75 mm - 0-10 0-5
Table 3.2 Limits of Coarse Aggregate Properties

Test Limit
Material Finer than 75m, by Washing (% maximum) 2

Absorption (% maximum) 2
Unconfined Freeze-Thaw Loss (% maximum) 6
Flat and Elongated Particles (% maximum) 20
Petrographic Number (maximum) 125
Micro-Deval Abrasion (% maximum loss) 14
Accelerated Mortar Bar (% maximum at 14 days) 0.15
27
Test Limit
Potential Alkali-Carbonate Reactivity of Quarried Carbonate Rock Non-Expansive
Concrete Prism Expansion (% maximum at 1 year) 0.04
1. 1% for gravel
Table 3.3 Fine Aggregate Gradation Limits

Sieve Size Percent Passing
9.5 mm 100
4.75 mm 95-100
2.36 mm 80-100
1.18 mm 50-85
600 m 25-60
300 m 25-60
150 m 0-10
75 m1 0-3
1. 0-6% for manufactured sand
Table 3.4 Limits of Fine Aggregate Properties

Test Limit
Organic Impurities (maximum organic plate number) 2
Micro-Deval Abrasion (% maximum loss) 20
Accelerated Mortar Bar (% maximum at 14 days) 0.15
Concrete Prism Expansion (% maximum at 1 year) 0.04
3.2.2 Aggregate Test Results
Coarse and fine virgin aggregate testing was conducted at Dufferin Aggregates as part of their on
going quality assurance (QA)/quality control (QC) process. Tests conducted in preparation for
mixture proportioning included: absorption, bulk relative density, and gradation for all aggregates.
Coarse RCA was also tested for crushed particles, flat and elongated particles, and micro-deval to
better characterize the material properties. Figures 3.2 3.13 present the aggregate test results. The
virgin aggregate is described by the results of the last 20 QA/QC periods as shown by the green
28
squares and blue diamonds. The MTO acceptance limit is shown as a solid line, and the RCA value is
shown as the gray triangle at the right of the figure when available.
Figure 3.2 Materials Finer than 75 m Test Results
Figure 3.3 Absorption Test Results
29
Figure 3.4 Bulk Relative Density Test Results
Figure 3.5 Crushed Particles Test Results
Figure 3.6 Flat and Elongated Test Results
30
Figure 3.7 Unconfined Freeze-Thaw Loss Test Results
Figure 3.8 Petrographic Number Test Results
Figure 3.9 Micro-Deval Abrasion Loss Test Results
31
Figure 3.10 Organic Impurities Test Results
Figure 3.11 Combined 37.5 mm and 19.0 mm Virgin Coarse Aggregate Gradation
Figure 3.12 19.0 mm minus RCA Coarse Aggregate Gradation
32
Figure 3.13 Fine Aggregate Gradation
Using the MTO OPSS virgin aggregate standards, the coarse RCA would have been rejected
because it exceeded the allowable limits for absorption, material finer than 75 m, and Micro-Deval.
3.3 Cement and Slag
The cement used for all concrete mixes in this research was St. Marys Cement General Use (GU) also
known as Type 10. This cement is 53% tricalcium silicate (C3S), 18% dicalcium silicate (C2S), seven
percent tricalcium aluminate (C3A), and eight percent tetracalcium aluminoferrite (CA4F) [St Marys
Cement 2007]. As given by the manufacturer, the Blaine fineness is 410 m2/kg and the compressive
strength gain of neat paste is shown in Figure 3.14.
Figure 3.14 Compressive Strength Development of Neat Paste
33
St. Marys Cement ground granulated blast furnace slag (GGBFS), with a Blain fineness of 532
2
m /kg, was used for all concrete mixes in this project. The benefits of GGBFS include that it has
cementious properties, lowers the heat of hydration, increases durability and improves long-term
strength.
3.4 Air Entrainment
Air entrainment provides freeze-thaw (F/T) resistance through the creation of a microscopic air void
system distributed throughout the concrete. It also improves the fresh properties of the concrete mix
by increasing workability, reducing segregation, and reducing bleeding. Euclid Airex-L was used in
all mixes at a dose of 55 ml/100 kg of cement to achieve adequate air entrainment.
3.5 Water Reducer
Reducing the water in concrete using a water-reducing admixture is beneficial by providing more
lubrication allowing for easier handling and finishing, increased strength and durability, and reduced
shrinkage and permeability. Euclid Eucon WR was used in all mixes at a dose of 250 ml/100 kg.
3.6 Open Graded Drainage Layer
The field test sections had an asphalt stabilized open graded drainage layer (ODGL), design by
Construction Testing Asphalt Lab Ltd. The asphalt contained PG 58-28 asphalt cement (AC) at
1.8%, Hot-Laid (HL) 8 stone with 1.07% absorption resulting in a bulk relative density of 2684
kg/m3. The gradation of the asphalt stabilized open graded drainage layer are based on MTO Special
Provision No. 313F07 and is shown in Figure 3.15. The primary purpose of the ODGL is to allow
moisture that has infiltrated the pavement to drain away from the structure.
Figure 3.15 HL8 Gradation for Asphalt Stabilized Open Graded Drainage layer
34
3.7 Concrete Mixture Design
The CPATT test track is located at the Region of Waterloo Waste management Facility in Waterloo,
Ontario, Canada. Waterloo is in the Southern part of Ontario, approximately three hours from the
Windsor Detroit border. It is located in a wet freeze zone with typical annual frost depths of 1.2 m.
Figure 3.16 shows the average daily temperature and the average monthly precipitation from 1998 to
2006.
Figure 3.16 Waterloo Climograph
The weather conditions at the test track produce a freeze-thaw environment. The Canadian
Standards Association (CSA) recommends that plain concrete in this type of freeze-thaw environment
should meet the guidelines for an exposure level of C-2. The C-2 exposure level is defined as a plain
concrete (no structural reinforcement) that is exposed to both chlorides and freezing/thawing [CSA
2000]. Based on the exposure level of C-2 the concrete must meet CSA A23.1 standard that includes
the follow criteria [CSA 2000]:
A water cement ratio less than 0.45,
A 28-day compressive strength greater than 32 MPa,
A maximum aggregate size of 14 to 20 mm, and
An air content of 5% to 8%.
The American Concrete Pavement Association (ACPA) further recommends [ACPA 1993]:
A cement content greater than 335 kg/m3,
A slump less than 100 mm,
A well-graded aggregate, and
Use of a curing compound.
35
Both the CSA and ACPA criteria were used as guidelines for optimzing the concrete mixture
design. However, in order to increase the durability of concrete pavements for highway application,
the MTO OPSS 1002 recommends increasing the aggregate to a maximum size of 37.5 mm. The
MTO OPSS 1350 also recommends replacing 25% of the cement content by mass with GGBFS to
increase durability and reduce costs.
As per the CSA standard, the final mix should achieve a compressive strength value of 32 MPa at
28 days. In order to reduce costs, the lowest cement content that reaches 32MPa at 28 days will be
used even if it is less than the 335kg/m3 recommended by ACPA. Three cement contents were
chosen for initial testing 315 kg/m3, 330 kg/m3, and 345 kg/m3.
3.8 Concrete Mixture Proportioning
The mixture components were calculated using the absolute volume proportioning method. The
aggregate was proportioned as per MTO OPSS using 65% coarse and 35% fine aggregate. Table 3.5
presents the material characteristics used in the mixture proportioning.
Table 3.5 Material Properties Used in Concrete Mixture Proportioning

Material BRD1 Absorption Moisture Fineness
3
(kg/m ) (%) Content (%) Modulus
37.5 mm Virgin Coarse Aggregate 2641 1.47 1.2 -
19.0 mm Virgin Coarse Aggregate 2713 1.47 2.7 -
19.0 mm RCA Coarse Aggregate 2379 4.41 1.4 -
Virgin Fine Aggregate 2683 1.16 0.34 2.6
Cement, General Use 3140 - - -
Slag, Type S 2920 - - -
Water 1000 - - -
Air Entrainment
Water Reducer
1. BRD = Bulk Relative Density
3.8.1 Proportioning Example
The following is an example of the mix proportioning for the test mix with 15% RCA Coarse
Aggregate and a 315 kg/m3 total cement content. The same mixture proportioning procedure is
followed for all test mixes.
36
Cement Content
Total cement content 315 kg/m3
Cement = ( 315 " 0.75) = 236 kg /m 3
Slag = (315 " 236) = 79 kg /m 3
Volume Required
! " 236 % 3
Cement =$ ' = 0.075 m
! # 3140 &
" 79 % 3
Slag =$ ' = 0.027 m
# 2920 &
!
Water Content
!
0.42 w/c
Water = (0.42 " 315) = 132 kg /m 3
Required Volume
" 132 % 3
Water =$ ' = 0.132 m
! #1000 &
Air Content
!
Required Volume
" 6 % 3
Air =$ ' = 0.06 m
#100 &
Aggregate Content
!
Total Volume
Aggregate = (1" (0.075 + 0.027 + 0.132 + 0.06)) = 0.706
Percent of Aggregate Requires
37.5 mm Virgin = (0.65 " 0.4 ) "100% = 26.0%
!
19.0 mm Virgin = (0.65 " 0.45) "100% = 29.25%
19.0 mm RCA = (0.65 " 0.15) "100% = 9.75%
!
Sand = (1" (0.26 + 0.2925 + 0.0975)) #100% = 35.0%
!
!
!
37
Volume of Aggregate Required
37.5 mm Virgin = (0.706 " 0.26) = 0.184 m 3
19.0 mm Virgin = (0.706 " 0.2925) = 0.206 m 3
19.0 mm RCA = (0.706 " 0.0975) = 0.069 m 3
!
Sand = (0.706 " 0.35) = 0.247 m 3
!
Mass of Aggregate !
37.5 mm Virgin
! = (0.184 " 2641) = 485 kg
19.0 mm Virgin = (0.206 " 2713) = 560 kg
19.0 mm RCA = (0.069 " 2379) = 164 kg
!
Sand = (0.247 " 2683) = 663 kg
!
Total = 1872 kg
!
!
Estimated Concrete Density
!
Assuming SSD Condition of Aggregates
Water = 132 kg /m 3
Cement = 315 kg /m 3
37.5 mm Virgin = ( 485 "1.0147) = 492.13 kg /m 3
!
19.0 mm Virgin = (560 "1.0147) = 568.23 kg /m 3
!
19.0 mm RCA = (164 "1.0411) = 170.74 kg /m 3
!
Sand = (663 "1.0116) = 670.69 kg /m 3
!
Total ! = 2348.79 kg /m 3
!
Moisture Corrections
!
37.5 mm Virgin = ( 485 "1.012) = 490.82 kg
19.0 mm Virgin = (560 "1.027) = 575.12 kg
19.0 mm RCA = (164 "1.014 ) = 166.30 kg
!
Sand = (663 "1.0034 ) = 665.25 kg
!
Total = 1897.49 kg
!
! 38
!
Water To Be Added
37.5 mm Virgin = (0.0147 " 0.012) # 485 = 1.31 kg
19.0 mm Virgin = (0.0147 " 0.027) # 560 = "6.89 kg
19.0 mm RCA = (0.04411" 0.014 ) #164 = 4.93 kg
!
Sand = (0.0116 " 0.0034 ) # 663 = 5.43 kg
!
Water = 132 " (1.31" 6.89 + 4.93 + 5.43)
!
= 137.08 kg
!
!
Batch Weights
!
Water (to be added) = 137.08 kg
Cement = 236.0 kg
Slag = 79.0 kg
!
37.5 mm Virgin = 490.82 kg
!
19.0 mm Virgin = 575.12 kg
!
19.0 mm RCA = 166.30 kg
!
Sand = 665.25 kg
!
Total = 2349.57 kg
!
!
3.9 Testing Method !
Laboratory testing was carried out on twelve test mixes in order to determine the best mixes to place
on the CPATT test track. A four by three factorial design was used consisting of four coarse RCA
amounts (0%, 15%, 30% and 50%) and three cement contents (315 kg/m3, 330 kg/m3, and 345
kg/m3). The mixes were proportioned using an absolute volume design method. Table 3.6
summarizes the quantity of each material used in the test mixes based on an aggregate saturated
surface dry condition. Values presented in the tables have been adjusted to reflect the moisture
content of the aggregates.
Figure 3.17 shows the combined aggregate grading for the four RCA mixes. A ratio of 65% coarse
aggregate to 35% fine aggregate was used.
39
Table 3.6 Mixture Proportioning Values
Mix Cement Slag Water 37.5mm 19mm 19mm Sand Air WR
(kg) (kg) (kg) Vir. (kg) Vir. (kg) RCA (kg) (kg) (ml) (ml)
1 236 79 130 491 767 - 665 173 788
0% RCA
2 247 83 136 482 754 - 654 182 825

3 259 86 143 475 742 - 644 190 863
1 236 79 137 491 575 166 665 173 788
15% RCA
2 247 83 143 482 566 163 654 182 825

3 259 86 150 475 557 161 644 190 863
1 236 79 144 491 383 332 665 173 788
30% RCA
2 247 83 150 482 377 327 654 182 825

3 259 86 157 475 371 321 644 190 863
1 236 79 154 491 128 553 665 173 788
50% RCA
2 247 83 160 482 126 544 654 182 825

3 259 86 166 475 124 536 644 190 863
Figure 3.17 Waste Concrete and Recycled Aggregate Combined Gradation

The compressive strength development at 28 days will be used to select a cement content that is
adequate for placement at the CPATT test track. For each of the twelve test mixes eight cylinders
were prepared for compressive strength testing. Compressive strength testing was conducted at five,
seven, fourteen, and 28 days. One of the two fourteen-day cylinders was used to conduct a hardened
air void analysis. Fresh properties including unit weight, slump, and air content were measured to
assess the mix for compliance with mix design.
40
3.9.1 Batching Method
The double mixing method is used to create areas of low w/c paste around the aggregates to improve
the strength and durability properties of the concrete. The concrete was batched using a modification
of the double mixing method. The double mixing method was modified to separate the addition of
the coarse and fine aggregates, and of the total water was added at each stage. The mixing steps
are outlined below:
1. Proportion all coarse aggregate and dry mix
2. Add water (1/4) and mix
3. Add proportioned fine aggregate and mix
4. Add water (1/4), AEA, WRA, and mix
5. Add cement and slag and mix
6. Add remaining water (1/2)
7. Mix and discharge,
A total of 0.06 m3 (2.12 ft3) of concrete was made for each of the twelve test mixes to perform
fresh property and compressive strength testing. The concrete was mixed in a portable drum mixer.
Each mix combination was batched in two 0.03 m3 quantities. The first batch was mixed and the
fresh properties of unit weight, air content, and slump were measured. Then, the first batch sat while
the second batch was mixed. Based on the fresh properties of the first batch, adjustments were made
to the amount of water in the second batch. The batches were then combined and mixed for
approximately three minutes. The fresh properties of the combined mix were taken and cylinders
were cast. For each of the twelve mixes, eight cylinders were cast in order to test for compressive
strength and perform an air void analysis. The batching was performed on four separate days; 0%
RCA mixes on May 31, 2007, 15% RCA mixes on June 1, 2007, 30% RCA mixes on June 4, 2007,
and 50% RCA on June 5, 2007.
3.10 Results
3.10.1 Plastic Properties
Unit weight, air content, and slump were measured on the plastic concrete of each of the combined
mixes. The unit weight was determined using CSA A23.2-6C (Density, Yield, and Cementing Factor
of Plastic Concrete) and a vibrator for compaction. Air content was measured using CSA A23.2-4C
(Air Content of Plastic Concrete by the Pressure Method). CSA A23.2-5C (Slump of Concrete) was
the procedure used to measure slump. The results of the plastic concrete testing are presented in
Table 3.7. All test mixes met the design criteria for an air content of 5%-8% and a slump of less than
100mm.
41
Table 3.7 Plastic Properties
Mix Unit Weight Air Content Slump
3
(kg/m ) (%) (mm)
1 2350 5.9 40
0% RCA
2 2367 5.6 40
3 2338 6.1 40
1 2325 6.8 60
15% RCA
2 2340 5.8 40
3 2338 5.8 40
1 2323 6.1 40
30% RCA
2 2315 6.0 40
3 2321 6.1 40
1 2306 5.7 40
50 %RCA
2 2301 6.1 40
3 2293 5.8 40
3.10.2 Hardened Properties
3.10.2.1 Air Content and Spacing Factor
Air void analysis was used to confirm the air content and spacing factor for the test mixes containing
315 kg/m3. Table 3.8 shows the results of the analysis and compares the air content of the plastic and
hardened concrete. The air content and spacing factor (L) were determined using ASTM C 457
(Standard Test Method for Microscopical Determination of Parameters of the Air-Void Content and
Parameters of the Air-Void System in Hardened Concrete).
Table 3.8 Air Content and Spacing Factor

Mix Air Content (%) Spacing Factor
Plastic Harden (mm)
0% RCA 5.2 5.9 0.134
15% RCA 6.6 6.8 0.180
30% RCA 5.3 6.1 0.125
50% RCA 5.6 5.7 0.143
42
The test mixes had an acceptable air content between 5-8%, and a spacing factor less than 0.23
mm. Since the mixes contain an acceptable air-void system they should provide sufficient freeze-
thaw protection.
3.10.2.2 Compressive Strength
Tests for compressive strength followed CAS A 23.2-9C (Compressive Strength of Cylindrical
Concrete Specimens). Compressive strength results are presented for each of the four coarse RCA
contents in Figures 3.18 3.21. The results presented are based on the average results of two
cylinders.
Due to equipment availability, the specified 5-day testing of the 0% RCA and 50% RCA was
shifted to day six and the 15% RCA to day four. All test mixes except the 30% RCA 315 kg/m3
reached 32 MPa compressive strength by fourteen days. All test mixes reached the design strength by
28-days. Compressive strength increased as cement content increased for all RCA mixes; however,
the opposite was observed for the 0% RCA. Figure 3.22 shows the 28-day compressive strength for
the varying RCA amount based on cement content.
Figure 3.18 Compressive Strength Development 0% Coarse RCA
43
44
Figure 3.22 Compressive Strength at 28 Days
3.10.3 Selection of Mix
Since all mixes reached the specified 30 MPa at 28-days, the mixes with the smallest amount of
cement content 315 kg/m3 were selected for field placement at the CPATT test track. Although the
cement content is lower than the ACPA recommended 335 kg/m3, adequate strength was reached and
it provides a more cost effective mix design.
3.11 Statistical Analysis
Two factor ANOVA analysis was performed to assess the significance of RCA content, cement
content and any interaction on compressive strength. ANOVA was conducted at a 95% confidence
interval with 2 replicates. The null hypothesis (Ho) for each of the three ANOVA replicates are as
follows.
Ho1: There is no effect of RCA content on 28-day compressive strength.
Ho2: There is no effect of cement content on 28-day compressive strength.
Ho3: There is no interaction between RCA content and cement content on 28-day compressive
strength.
The ANOVA results are summarized in Table 3.9.
Since F calculated is greater than F critical for each analysis, all three null hypothesis are rejected.
Therefore the RCA content, cement content, and their interaction are statistically significant in
explaining the observed variance in the resulting 28-day compressive strength.
45
Table 3.9 Results of ANOVA
Source SS df MS F F critical
RCA 29.15 3 9.72 5.13 3.03
Cement 53.59 2 26.80 14.16 3.42
Interaction 73.89 6 12.32 6.51 2.53
Error 22.71 12 1.89
Total 179.34 23
46
Chapter 4
Pavement Design
Prior to the design of the test track sections, information was gathered on the existing field conditions
of the soil subgrade and granular base as well as traffic patterns and loads. Based on this information,
the thickness and jointing strategy of the jointed plain concrete pavement surface (JPCP) was chosen.
In order to compare the long-term durability of JPCP containing RCA and conventional JPCP,
sensors for strain, slab curling and warping, joint movement, maturity and temperature were included
in the test track design.
4.1 Soil Investigation
MacLaren Engineering Ltd. conducted a soil investigation in the test track area on three occasions
(February 1983, December 1993, and August 1995). Boreholes were drilled within the right-of-way
of the future asphalt test sections. Results of the soil investigations are presented in Tables 4.1 to 4.3.
The subgrade was classified as clayey silt sand and gravel with moist conditions. The standard
penetration resistance N values ranged from 18 to 50+ blows per 0.3 meters which is a medium to
very dense soil classification [Craig 2007].
Table 4.1 Borehole Log OW7-83

Depth (m) Description
0.0 - 1.7 Grey brown clay, fine sandy silt
1.7 - 3.2 Grey brown fine sand silt
3.2 - 4.7 Grey brown silty clay till
4.7 - 6.2 Grey clay silt till
6.2 - 12.2 Grey silt clay till
12.2 - 13.9 Brown silt and fine to medium sand
13.9 - 16.9 Mottled brown and dark brown clay, fine cand and silt
16.9 - 18.4 Mottled brown and dark brown silty clay
18.4 - 20.0 Mottled brown and dark brown clayey silt
47
0.0 - 0.9 Brown and grey fill
0.9 -3.8 Brown grey clay silt till
3.8 - 11.4 Grey clay silt till
11.4 - 14.5 Brown grey clay silt till
14.5 - 20.6 Brown grey fine san with medium sand and silt

0.0 - 3.7 Brown clay silt with traces of sand
3.7 - 5.2 Brown silty sand till with traces of gravel
5.2 - 6.7 Brown sans silt till with traces of clay and gravel
6.7 - 9.0 Interceded layers of brown silt and silty sand tills
9.0 - 12.7 Brown silt and sand
4.2 FWD Testing
Stantec Consulting Ltd conducted Falling Weight Deflectometer (FWD) testing of the granular base
material on April 25, 2007 prior to paving. There were twelve tests conducted in total, the 15% and
50% RCA sections had two FWD test sites and the 0% and 30% RCA sections had one FWD test site.
There were six test sites in each of the Northbound (NB) and Southbound (SB) lanes. FWD testing
was conducted using a Dynatest Heavy Weight Deflectometer (HWD) as shown in Figure 4.1.
Figure 4.1 Dynatest HWD
48
Table 4.4 shows the Dynatest HWD spacing of the nine geophones from the applied load.
Table 4.4 Geophone Sensor Configuration (mm)

D1 D2 D3 D4 D5 D6 D7 D8 D9
0 300 450 600 900 1,200 1,500 1,800 -300
At each of the twelve test locations, four different loads were applied. A seating load ensured
contact between the load plate and the pavement surface, then 40 kN (9,000 lb), 45 kN (10,115 lb),
and 60 kN (13,500 lb) were applied. Table 4.5 presents the FWD results.
Table 4.5 FWD Results

RCA D1 (m) MR (MPa) Ep SN
Amount NB SB NB SB NB SB NB SB
0 1,144 1,435 36.7 28.7 111,587 89,500 2.049 1.903
15 1,240 1,376 41.5 37.1 103,918 88,190 1.986 1.894
30 1,118 2,000 88.1 72.3 95,723 50,886 1.947 1.577
50 1,206 1,358 58.6 71.8 140,699 106,164 2.135 1.960
D1 sensor detected a larger deflection in the SB direction then in the NB direction. The largest
deflection was the 30% RCA section in the SB direction indicating the weakest area of granular and
subgrade. Other than the 30% RCA SB section all other sections gave consistent results.
Resilient modulus (MR) is an indication of the subgrade elasticity under loading. An MR < 31 MPa
(4500 psi) is identified as a low or weak area. [AASHTO Design Guide Part 2, Chapters 1,2] There
were variable results for both NB and SB directions. The 0% RCA section in the SB direction is
classified as a weak section whereas the other sections were in the acceptable range. Using MR,
California Bearing Ratio (CBR) was calculated using the following equation [TAC 1997].
M R ( MPa) = 10.3 CBR (5.1)
CBR values range from 2.78 to 8.55 with an average of 5.27 and is representative of a clay
material. !
Ep measures the elasticity under loading of the aggregate base material. In general, the NB has a
higher Ep demonstrating a faster return to its original shape after a load. The 30% RCA section had
the lowest Ep.
49
The pavement structural number (SN) was consistent for both the NB and SB directions except for
30% RCA. This shows again that the 30% RCA SB section is the weakest. Using the following
equation and the SN results from the FWD the layer coefficients can be calculated.
SN = a1D1 + a2 D2 m2 + a3 D3 m3 (5.2)
where: a1, a2 , and, a3 = layer coefficients of the surface, base, and subbase layers
!
D1, D2 , and, D3 = thickness of the surface, base, and subbase layers (inches)
m2 , and m3 = drainage coefficients of the base and subbase layers
!
!
Assuming a uniform granular layer of 450 mm (17.75 in) for each test sections, the layer
! coefficient is between 0.09 and 0.12. These values are representative of a crushed stone and are
consistent with what is observed in the field.
4.3 Site Survey
CPATT conducted a site survey on April 17, 2007 to determine the centerline profile of the existing
granular base. The length available to construct the four test sections is 172m. This was measured
from the edge of existing asphalt to the start of the right-hand turn guardrail. The elevation decreased
5.15 m over the 172 m length of the test sections. Figure 4.2 shows the base profile along the
proposed centerline.
Figure 4.2 Test Section Profile
50
4.4 Estimated Traffic Load
Equivalent Single Axle Load (ESALs) on the test track sections was calculated using equation 4.3.
This equation was developed for rigid pavement during the original AASHO road test [AASHTO
1993].
4.62
" G %
Wx " L18 + L2s % $ 10 ( x ' 3.28
=$ ' $ G '( L2x ) (4.3)
W18 # Lx + L2s & $ (18 '
#10 &
Wx
where: = axle applications
! inverse of equivalency factors
W18
Lx = axle load being evaluated (kips)
L18 = standard axle load based on 18,000 lb (80 kN) single axle load (kips)
!
L2 = code for axle configuration
!
1 = single axle
!
2 = tandem axle
!
3 = triple axle
x = axle load equivalency factor being evaluated
s = standard single axle
# 4.5 " pt &
G = log% (
$ 4.5 "1.5 '
pt = terminal serviceability index
# 3.63( L + L ) 5.2 &
x 2x
! " = 1+ %% 8.46
(
3.52 (
! $ ( D + 1) L2x '
D = slab depth (in)
!
A Monte Carlo simulation, with 25,000 iterations, was used to estimate the ESAL value. The
!simulation was performed using two design vehicles, a Region of Waterloo standard garbage truck
and an articulated dump truck. The traffic volume was simulated using 54,000 and 7,500 for the
garbage and dump trucks respectively.
51
The loaded range of the garbage and articulated dump trucks are shown in Table 4.6 [Sterrling
Acterra 2006, CAT 2006]. In the Monte Carlo simulation, a uniform distribution of axle loading was
assumed. The loaded condition ranged from 75% to 110% of the manufacturers maximum load.
Table 4.6 Manufacturer Stated Axle Loads

Axle Garbage Truck Articulated Dump Truck
(tonnes) (lbs) (tonnes) (lbs)
Sterring 9.07 20,000 15.87 34,965

Tandum 20.87 46,000 35.10 77,404
Figure 4.3 shows the results of the Monte Carlo simulation run for the design year. Using a 95%
confidence level the upper limit of the ESAL calculation is 1,510,000.
Figure 4.3 Monte Carlo Simulation Results
4.5 Sensors
Sensors were used to measure both static and dynamic conditions. Static measurements included
longitudinal and transverse strain in the concrete, displacement of concrete (slab curl/warp, shrinkage,
joint movement), concrete maturity, and temperature, moisture and humidity of the pavement layers.
Sensors were needed to measure dynamic conditions of load, vertical displacement, and longitudinal
and transverse strain in the concrete. Sensors needed to measure static conditions every one to two
hours and dynamic conditions for each loading event.
Some key considerations in selecting sensors were taken from published literature [Farhey 2007].
The measurement length of the sensor is proportional to the measurement range.
52
If possible, embedment of sensors in the pavement should be avoided since it can
potentially cause some micro-cracking (stiffnesssensor > stiffnessmaterial) or the formation of
voids (stiffnesssensor < stiffnessmaterial) in the pavement causing a change in the systems
behaviour.
If embedment sensors are necessary, then choose sensors that have an unobtrusive shape
and a distributed anchor system.
Incorporate sensor redundancy as it is likely that the number of sensor failures will increase
as the length of the test increases.
Avoid sensors that require recalibration to correct measurement drift (i.e. fiber optic) for
long-term monitoring.
Meetings were held with the distributor, Hoskins Scientific Limited, and the sensor manufacturer,
RST Instruments Ltd, to select instrumentation for the test sections. Information was gathered on the
type of sensor required for each type of measurement.
Strain is measured using multiple vibrating wire embedment sensors.
Moisture content is measured with time domain reflectometry (TDR) probes. There was
concern about TDR probes being placed in concrete because of the alkaline pore solution
reacting with the metal waveguides. A previous case in the literature showed that using a
coated 20 cm waveguides TRD probes could accurately determine the actual moisture
content of the cement without loss of refraction. The coating protected the TDR from the
high alkaline content [Soilmoisture 1998].
Curling and warping is measured with multiple vibrating wire embedment sensors and/or a
tilt meter.
Joint movement is measured using a vibrating wire embedement.
Displacement sensors for curling/warping and joint movement should have the stroke set in the
middle to allow for both positive and negative movements. Separate systems are required to monitor
static and dynamic measurements.
The first sensor design proposed included nine strain gages, three vertical extensometers to
measure curling and warping, three inter-panel extensometers to measure joint movement and five
TDR probes to measure moisture content. One JPCP slab was instrumented in each of the four test
sections.
Each instrumented slab would have sensor cables running to the shoulder of the road connecting
into a remote multiplexer, each multiplexer would need to contain three Flexi-Mux to accommodate
the fifteen vibrating wire sensors. The desired sensor cable length is ten meters. The TDR cables are
not compatible with the remote multiplexer so the cables would be stored in a small enclosure on the
shoulder of the road for access at specific reading times. Each test sections remote multiplexer would
be connected in series with the datalogger being located in the middle of the RCA sections. Prior to
concrete placement, sensor cables would be placed inside PVC conduit that ran from the sensor
53
through the asphalt stabilized OGDL layer, into the granular material 0.3 m (1 feet) and out to the
road shoulder. Figure 4.4 is a diagram of the sensor conduit.
Three conduits were required for each slab. One conduit contained the vertical extensometer
cables along the south end of the instrumented slab. A second conduit, located mid-slab, housed the
cable for the strain gages and maturity sensors. A third conduit ran along the north side of the
instrumented slab and contained the cable from the inter-panel extensometers and strain gages.
Figure 4.4 Sensor Conduit

Protecting the sensors from the slip form paver during placement of the concrete was of utmost
importance. The strain gages were put on chairs with a solid base that could be anchored into the
asphalt using dowel basket pins. The gages were mounded with concrete to protect them from the
paver. The bottom of the vertical extensometer was attached to rebar anchored in the granular layer.
It was mounded with concrete to protect them from the paver. The inter-panel extensometers were
placed inside a blockout during paving for protection.
Due to cost restrictions, the first sensor design that was proposed was revised. The final layout of
the sensors included twelve sensors in each test section. The sensors used in the revised design
included six strain gauges, two vertical extensometers, two inter-panel extensometer and two maturity
sensors. These twelve sensors were duplicated in each of the four sections (0%, 15%, 30%, and 50%
RCA). The sensors removed from the first design are 3 strain gauges at slab mid-depth, and one each
of the vertical and inter-panel extensometers along the mid-slab joint. Figure 4.5 shows the revised
location of all the sensors.
Figure 4.5 Revised Sensor Locations
54
4.6 Design Thickness
The American Concrete Pavement Association (ACPA) StreetPave design software was used to
design the thickness of the JPCP. The entries for global settings and project level inputs were:
climatic region mean annual air temperature (MAAT) of 7.2 o C (45 oF),
terminal pavement serviceability index (PSI) of 2.25 to represent arterial and collector
roads,
15% cracked concrete slabs at the end of the design life to represent arterial and collector
roads,
design life of 35 years with reliability set at 90%.
Since the actual traffic patterns may vary from the estimate, the StreetPave traffic settings were set
to slightly over estimate the traffic. The traffic category was set as a major arterial road. The average
annual daily truck traffic of 5000 was weighted evenly between the two lanes. A 4% traffic growth
rate was used.
Pavement settings were chosen based on representative average values. A California Bearing
Ratio (CBR) of five was used to represent the average field condition of clay. The compacted
granular material was 450 mm with an MR of 200 MPa and the asphalt stabilized OGDL was 100
mm with an MR of 1200 MPa. The combined values give a composite modulus of subgrade reaction
(k) equal to 148.2. The average 28-day flexural strength (MR) was set at five megapascals and the
concrete modulus of elasticity (E) was set at 33,750 MPa.
StreetPave results indicated that the minimum thickness for the JPCP layer was 199.9 mm. The
recommended thickness was 203.2 mm with a maximum transverse joint spacing of 4.27 m and a
dowel bar diameter of 31.75 mm.
It was decided to use a pavement thickness of 250 mm to accommodate the cover requirements of
the automatic dowel bar inserter, reduce the likelihood of vibrator trails from the slipform paver,
allow for comparison of performance with MTO standard JPCPs, and to provide opportunity for
multiple diamond grindings to restore the pavements smoothness as required.
4.7 Preconstruction Meetings
In preparation for the placement of the test sections, six preconstruction meetings were held.
Representatives from the University of Waterloo, the CAC and Dufferin attended. Summaries of the
meetings are in Appendix B.
55
Chapter 5
Construction
5.1 Granular Base
The Region of Waterloo had previously constructed the granular base to use as the South access haul
road for the proposed SE-1 and SE-3 landfill expansion cells. On June 5, 2007, prior to paving,
Dufferin Construction preformed the final grading and compaction to bring the granular base
structure to the desired profile.
5.2 Sensor Conduit
On June 9, 2007, conduits for the sensors were placed. The University of Waterloo laid out the test
track sections by marking the location were paving is to start and the location of joints. The locations
of the four slabs to be instrumented were identified and the location for the sensor conduits was
marked as per Figure 4.4.
Each of the four slabs required 18.5 m of ABS pipe, 0.75 m of PVC pipe, three elbows, and three
tees for the sensor conduits. Trenches for the sensor conduits were initially dug with pick-axes and
shovels. Due to the compaction of the granular base, a jackhammer was obtained to help breakup the
surface. Figure 5.1 pictures the excavation of the granular layer for sensor conduit placement.
Figure 5.1 Digging Sensor Conduit Trenches

The sensor conduit trenches were excavated to a depth of 305 mm (12 in). A 38 mm (1.5 in) ABS
pipe burrier was used in the granular base and 38 mm PVC pipe was used to extend up through the
base 100 mm to allow access to the conduit after the asphalt stabilized OGDL was placed. PVC pipe
was used in the OGDL since it has a higher melting point, 212oC, compared to ABS pipe, 105 oC.
The three sections of pipe were cut to the desired lengths allowing for the necessary couples, tees, and
elbows to be added. Then a thin layer of bedding sand was spread on the excavated trench to protect
56
the conduit from protruding aggregate. After the conduit pipe was place in the trench the area around
the conduit was backfilled with the excavated material in 50-76 mm (2-3 in) lifts and compacted with
a Marshall hammer. A small amount of water was added to the backfill to aid compaction. Figure
5.2 shows the compaction with a Marshall hammer of the granular material around the sensor conduit.
Figure 5.2 Marshall Hammer Compaction Around Sensor Conduit
5.3 Asphalt Stabilized Open Graded Drainage layer
Dufferin Construction requested that the PVC sensor conduit extensions be cut flush with the granular
base material so it would not interfere with paving. After the PVC conduit was cut approximately
10mm above the granular base, rags were stuffed into the conduit to prevent the asphalt stabilized
ODGL from plugging the opening. Paving was cancelled the night of June 10, 2007 due to a
breakdown at the plant that would not permit the manufacturing of the asphalt stabilized OGDL and
was rescheduled for June 11, 2007. On June 11, 2007 UW and Dufferin construction were onsite by
6:00 pm. The first load of asphalt stabilized OGDL was placed at 6:24 pm. The outside temperature
at the start of paving was 25 oC, and by the end of paving the temperature had dropped by six degrees
Celsius over four hours to nineteen degrees Celcius. There were clear skies with no precipitation.
A total of fourteen loads of asphalt stabilized OGDL, totaling 293.04 tonnes were used to pave
both lanes (8.5 m wide) of the 180m test sections. Table 5.1 shows the details of each load of OGDL
including the time the load was batched and placed, weight and location of placement. Figure 5.3
pictures the OGDL paving and compaction.
The test sections were paved one lane at a time. First the unloaded lane was paved then the loaded
side in the northbound direction (i.e. station 0+280 to 0+100). Paving depth was consistently between
90 mm and 115 mm. A two percent cross slope towards the edge of the pavement was created. The
OGDL was compacted using a vibratory asphalt compactor (CB-634D) over multiple passes and
started 30 minutes (7:00 pm) after the start of paving (6:30 pm).
57
Table 5.1 ODGL Load Details
Load Time Tonnes Location
Batched Placed Current Total Start Stop
1 17:24 18:24 20.93 20.93 NB 0+280 NB 0+260
2 17:40 18:37 21.02 41.95 NB 0+260 NB 0+245
3 17:56 18:44 20.93 62.88 NB 0+245 NB 0+220
4 18:14 18:55 20.95 83.83 NB 0+220 NB 0+200
5 18:28 19:30 20.94 104.77 NB 0+200 NB 0+170
6 18:43 19:37 20.94 125.71 NB 0+170 NB 0+139
7 19:13 19:53 20.88 146.59 NB 0+139 NB 0+120
8 19:37 20:18 20.87 167.46 NB 0+120 NB 0+100
9 19:54 20:47 21.01 188.47 SB 0+280 SB 0+253
10 20:06 20:52 20.80 209.27 SB 0+253 SB 0+230
11 20:34 21:25 20.94 230.21 SB 0+230 SB 0+197
12 20:55 21:39 20.93 251.14 SB 0+197 SB 0+172
13 21:14 21:55 21.02 272.16 SB 0+172 SB 0+138
14 21:54 22:36 20.88 293.04 SB 0+138 SB 0+100
Figure 5.3 Paving and Compaction of ODGL
The temperature of each load was measured and recorded at the time of placement. The
temperature was measured immediately after passing the screed of the paver and taken at mid-depth,
50 mm below the surface. The mid-depth temperatures ranged from 131.9 oC to 171.8 oC, with an
average of 148.6 oC, and a standard deviation of 8.8 oC.
58
A plate sample was taken from each test section location for laboratory testing with the following
details:
15%RCA sample from Load #3, temperature 158.0 oC
50% RCA sample from Load #7, temperature 152.7 oC
0% RCA sample from Truck #9, temperature 152.6 o C
30% RCA sample from Truck # 12, temperature 148.2 oC
The asphalt stabilized ODGL was allowed to cure 24 hours before concrete was placed.
On June 12, 2008, the openings to the buried conduit were located and clothesline was threaded
through the conduit to allow for faster placement of sensors during concrete paving.
5.4 Jointed Plain Concrete Pavement
Paving of the PCC layer was carried out over two days on June 13 and 14, 2007. Construction of the
JPCP started June 13, 2009. UW and Dufferin construction were onsite by 5:30 am and paving
started at 6:59 am. Concrete was delivered to the test section site in ready-mix trucks from the
Dufferins Kitchener Plant on Forwell Road, sixteen kilometers away from the site or a 25-minute
drive.
Paving was done using a slipform paver (Gunter and Zimmerman S850 Quadra) equipped with
auto float and burlap drag. Wood and magnesium floats on long poles were used on either side of the
road. Also, there was a hand finishing platform and a texture/cure machine. (Gomaco T/C 600)
Figure 5.4 shows the view of the paving setup from on top of the slipform paver.
Figure 5.4 Pavement Construction

The slipform paver had multiple starts and stops to allow for discharging the concrete from ready
mix trucks, sensor installation, and dowel basket installation. This produced localized low spots
where the paver was stopped.
A set of two dowel baskets consisting of thirteen 38 mm diameter epoxy coated dowel bars were
spaced every 300 mm with a height of 125 mm. The dowel baskets were placed using the variable
59
joint spacing sequence. Pins, 100 mm in length, were used to secure the load transfer devices to the
asphalt stabilized OGDL. The pins were secured on both sides of the load transfer device and placed
between every two to three dowel bars. The tie straps were cut and removed from the load transfer
device except for the first set where they were cut but left attached.
Under the supervision of Jai Tiwari from Dufferin Construction, Eugene Kim, Xin Xu, and Jodi
Norris from CPATT carried out concrete fresh property testing and specimen preparation. The fresh
properties that were tested included air content, slump, temperature and unit weight and are shown in
Table 5.2. Unit weight measurements were taken from the same load that was used to cast specimens
for each RCA amount.
Table 5.2 Concrete Fresh Properties

RCA Ticket Time Air Slump Unit Weight Temperature
3
Amount Number Tested Content (%) (mm) (kg/m ) (oC)
135147 6:21 5.6 20 - 22.5
0% 135151 6:45 6 30 2371 23.7
135155 7:30 7.2 50 - 23.8
135162 8:38 6.8 40 - 22.3
135165 9:05 6.1 50 - 22.9
15% 135168 9:55 5.9 50 2335 24.7
135173 10:40 7 40 - 24.7
135194 2:00 5.5 30 - 22.8
135198 2:30 5.3 50 - 23.6
135201 3:25 6.2 40 2316 24.4
30%
135206 4:10 6 40 - 23.2
135209 4:37 4.9 50 - 24
134213 5:15 5.5 40 - 22.9
130217 6:05 5.8 30 - 24.8
50% 127220 6:25 6.2 40 2298 24.3
121227 7:32 5.4 40 - 25.1
114233 8:30 6.8 40 - 23.6
60
A number of specimens were cast for each RCA amount. There were nine cylinders for
compressive strength testing, one cylinder for air void analysis, two cylinders for maturity testing,
three beams for flexural strength testing, and two beams for freeze-thaw testing. Due to low slump
values, a pencil vibrator was used to consolidate the concrete in the cylinders and beams as allowed
by ASTM C 231. The cast specimens were cured in the field for 24 hours before being transferred to
the CPATT lab at UW to be demolded and stored in the fog room until testing as per ASTM C31.
When paving started, the outside air temperature was 20.9oC with sunny clear skies and no
precipitation. The temperature increased 8.8 o C over four hours to a temperature of 29.7 o C at 11 am.
At 11 am the temperature of the asphalt stabilized ODGL was 52oC or higher. Due to the high air
temperature and weak compressive and tensile strength of the asphalt stabilized ODGL it would
deform when the trucks drove on it creating high and low spots and a non-uniform paving surface.
Water was used to cool the asphalt stabilized ODGL to help prevent moisture loss of the concrete and
reduce the deformations of the surface. However, the non-uniform surface of the asphalt stabilized
ODGL caused vertical skewing of the dowel bars as seen in Figure 5.5.
Figure 5.5 Skewing of the Dowel Bars

CSA A23.1 states that for regular PCC evaporation rates greater than 1.0 kg/m2, significantly
increases the potential for shrinkage cracking. By adding in pozzolans (fly ash or silica fume)
evaporation rates over 0.5 kg/m2 increase shrinkage cracking potential. Although the RCA mixes did
not contain any pozzolans, RCA is more susceptible to moisture loss, which also increases the risk of
shrinkage cracking. The evaporation rate at 10:45 am was estimated at 0.6 kg/m2 per hour. Using the
Cement Association of Canadas Design and Control of Concrete Mixtures, the evaporation rate was
calculated using an air temp of 29.3 oC, a relative humidity of 41.1%, a concrete temperature of 28.3
o
C, and a wind velocity of 8.5 km/h. Dufferin construction decided to stop paving at 11 am since the
concrete was becoming increasingly difficult to place and finish. Figure 5.6 illustrates the pavement
condition including a loss of pavement edge and voids in the surface.
61
Figure 5.6 Condition of JPCP
The remaining mixed concrete already onsite was placed (station 0+223 to 0+210) and would be
removed prior to paving the following day. As shown in Figure 5.7, plastic tarps were placed on top
of the asphalt stabilized ODGL for easier removal of the waste concrete. Dowel bars in the basket
were replaced with plastic tubes filled with sand to provide a reference of where the new bars would
be placed.
Figure 5.7 Placing Concrete for Removal

The paving that was completed the first day is summarized in Table 5.3 and the texturing
completed the first day is summarized in Table 5.4.
Table 5.3 Day 1 Paving

RCA NB (Unloaded) SB (Loaded)
Amount Starting Finishing Distance Starting Finishing Distance
Station Station (m) Station Station (m)
0% 0+280 0+253 27 0+280 0+258 22
15% 0+253 0+223 30 0+258 0+223 35
62
Table 5.4 Day 1 Texturization
Texturization Starting Finish Distance
Station Station (m)
Longitudinal burlap drag with 16 mm c/c longitudinal 0+280 0+256 24
tining
Longitudinal burlap drag only 0+256 0+223 33
UW and Dufferin construction were onsite by 11:00 pm on June 13. Prior to the start of paving,
concrete from station 0+223 to 0+210 was cut into smaller pieces and removed as shown in Figure
5.8.
Figure 5.8 Removal of Waste Concrete

Removal of the JPCP disturbed the underlying asphalt stabilized OGDL. A sixteen-millimeter
clear stone was used to replace the disturbed material without compaction. Dowel bar holes were
drilled with a single hand-held drill. The dowel holes were cleaned before installing the grout and
dowel bars. As shown in Figure 5.9, the reestablished joint prior to paving had some misalignment of
the dowel bars, and some voids between the grout and the dowel.
Figure 5.9 Reestablishing Joints Prior to Paving
63
Paving began around 1:30 am and the last load was placed by 9:55 am. The 30% RCA and 50%
RCA mixes were adjusted to include a Type R retarder to slow down the setting of the concrete, to
counteract the effect of high temperature and allow time to properly finish the concrete. When paving
started, the outside air temperature was 20.5 o C with clear skies and no precipitation. The temperature
increased 2.6 oC over 8.5 hours for a final temperature of 23.1 oC at 10 am. The highest temperature
of the asphalt stabilized ODGL was 39 oC.
Compared to the previous day, deformation of the asphalt stabilized ODGL layer continued but
was less significant due to the lower ambient temperature. The amount that the slipform paver was
stopped was reduced by running cable through the conduit ahead of time and creating a small trough
in the asphalt stabilized OGDL for the cables so they would not be pinched and the sensor would not
be run over. This allowed the ready-mix trucks greater access to the slipform paver.
The first application of curing compound was applied a maximum of one hour after the placement
of the concrete. Paving that was completed on the second day is shown in Table 5.5 and the texturing
completed is shown in Table 5.6.
Table 5.5 Day 2 Paving

RCA NB (Unloaded) SB (Loaded)
Amount Starting Finishing Distance Starting Finish Distance
Station Station (m) Station Station (m)
15% 0+223 0+205 18 0+223 0+210 13
30% 0+205 0+165 45 0+210 0+160 50
50% 0+165 0+100 65 0+160 0+100 60
Table 5.6 Day 2 Texturization

Texturization Starting Finishing Distance
Station Station (m)
tining
16 mm c/c longitudinal tining 0+180 0+149.5 30.5
Transverse broom drag with 16 mm c/c transverse 0+149.5 0+138 11.5
tining
Transverse broom drag with 16 mm c/c longitudinal 0+138 0+110 28
tining
tining
64
The thickness of the JPCP was measured every fifteen meters over the length of the test sections
for a total of eleven measurements. The thickness was measured to the nearest 5 mm. The thickness
ranged from 250 mm to 310 mm and had an average of 272 mm and a standard deviation of 19.7.
The thickness increased over the last two slabs approaching the asphalt to provide suitable depth for a
proper transition.
Over the next 28-days at the test track the temperature ranged from 6.5 oC to 32 oC with an average
temperature of 20.8 oC, a standard deviation of 3.5 oC and total precipitation of 37.8 mm.
5.5 Early Age Behaviour
During the first three days after placement, portland cement concrete pavements are particularly
susceptible to shrinkage induced cracking. This cracking can have critical effects on the long-term
performance of the pavement. Cracking occurs when the tensile stresses exceed the tensile strength
of the concrete. Shrinkage induced cracking is caused by a number of factors including material type,
mix design, pavement design, construction practices and climate conditions. The increased
absorption rate of recycled concrete aggregate makes it particularly sensitive to weather conditions.
Higher temperatures and wind will accelerate water loss from the concrete to the atmosphere or to the
asphalt stabilized open graded drainage layer. This water loss can cause inadequate moisture in the
concrete negatively impacting hydration, strength development and surface finishing.
Computer simulations were run using HIPERPAV II software [Transtec 2005]. HIPERPAV II uses
data on pavement design, concrete mix design, the weather, and construction method to calculate
stress development in the concrete and compares it to strength development over the first 72 hours.
This allowed potential times of failure with uncontrolled cracking to be identified and avoided before
the construction of test sections. The simulations identified the possible early age cracking of the
RCA concrete when the ambient temperature exceeded 30 oC during placement of the JPCP.
Inspection of the test sections after 72 hours showed no signs of uncontrolled cracking of the
concrete pavement. Additional HIPERPAV II simulations were run using weather data from the
onsite weather station to confirm the observations. Figures 5.10 to 5.14 shows the simulation results
for all the test sections.
Figure 5.10 HIPERPAV II Results 0% Coarse RCA, Day 1
65
66
The simulation results confirmed the field observations that no uncontrolled cracking would
develop. During the 72 hours the maximum tensile stresses to tensile strength ratio of the concrete
reached 0.74 for the 30% Coarse RCA test section.
5.6 Joints
Dowel baskets were placed according to the variable joint spacing 3.7 m, 4.5 m, 4.0 m, and 4.3 m in a
repeating pattern. The location of where the joints were to be cut was marked with spray paint using
a string-line. Joints were marked with variable offsets from the centre of the dowel baskets at 0 mm,
25 mm, 50 mm, 75 mm and 100 mm as per Figure 5.15.
Figure 5.15 Joint Offsets

The initial sawing of the concrete joints was started 6.5 hours after paving started at 1:30 pm.
Sawcutting was preformed by BASIC Concrete Cutting Inc. using a Husquvarna FS6600D concrete
cutting saw. The saw was equipped with a 3.175 mm (1/8 inch) diamond saw blade and was set to
cut a depth of a third of the thickness of the slab, 85 mm. The transverse joints were cut first
followed by the longitudinal joints. Figure 5.16 shows the initial saw cut.
67
Figure 5.16 Initial Saw Cuts
On July 7, 2007, Road Master Construction widened the original transverse and longitudinal joints
and sealed the joints. This joint sealing was a three-part process. First, the original cuts were
widened to a width of fifteen millimeters and depth of 35 mm to create a reservoir for the joint
sealant. The cuts were cleaned with dry sandblasting to clean away any cutting residue and add
texture to the reservoir walls for extra adhesion. Joints were cleaned a second time using air to
remove sand, dirt, and dust and a visual inspection of the joint was done to confirm it was clean.
Second, the nineteen-millimeter backer rod was inserted into the joint using a steel roller. Inspection
of the backer rod showed that it was cut or stretched by the roller during installation. Finally, asphalt
rubber based joint sealant was poured into the reservoir. Figure 5.17 shows the widening and sealing
of the transverse and longitudinal joints.
68
Figure 5.17 Widening and Sealing the Initial Saw Cut
During the widening of the joints, the edges of the pavement became damaged in several locations
as pictured in Figure 5.18.
Figure 5.18 Pavement Damage
5.7 Sensors
The maturity sensors arrived on June 4, 2007 and the strain gauges, vertical extensometer, and inter-
panel extensometer arrived on June 8, 2007. The datalogger, the four multiplexers and the solar panel
were delivered approximately two to three weeks after the installation. On June 12, 2007, the strain
gauges and maturity sensors were attached to the chair structure as shown in Figure 5.19.
69
Figure 5.19 Strain Gauge and Maturity Setup
Chairs allowed the strain gauges to be securely fastened into the asphalt stabilized ODGL and held
the strain gauges at 50 mm above the bottom of the JPCP and 50 mm below the top of the 250 mm
thick pavement structure. As seen in Figure 5.20, rebar was added to the vertical and inter-panel
extensometers to hold them in place.
Figure 5.20 Vertical Extensometer

To install the vertical extensometer, a pickaxe was used to create a small hole in the asphalt
stabilized ODGL to insert the sensor. This allowed 50 mm of JPCP to cover above the rebar.
Extracted asphalt material was placed around the sensor and gently compacted to hold it in place.
Figure 5.21 shows how concrete was mounded over the sensor to protect it during paving.
The chair mounted strain gauges and maturity sensors were fastened to the asphalt stabilized
OGDL using four dowel basket pins, one in each corner of the chair. As seen in Figure 5.22, concrete
was mounded over the sensor to protect it during paving.
70
Figure 5.21 Mounding the Vertical Extensometer with Concrete
Figure 5.22 Mounding the Stain Gauges and Maturity Sensors with Concrete
The inter-panel extensometer sensor was wrapped in plastic and placed inside a blockout. The
blockout was filled with dirt and anchored into the asphalt stabilized ODGL and the granular base. To
accommodate the blockout two sections of the dowel basket were removed. Figure 5.23 shows the
blockout and dowel basket prior to paving. Traffic cones were placed on top of the blockouts to make
them more visible to the ready mix trucks backing up.
Figure 5.23 Inter-panel Extensometer Blockout Before Paving
71
The data logger and the four multiplexers were installed on July 5 and July 6, 2007. The
multiplexers and datalogger boxes were installed on metal fence t-rails one to two meters from the
edge of the pavement as shown in Figure 5.24.
Figure 5.24 Location of Multiplexer and Dataloger Cabinets
The landfill management requested that the multiplexers be moved away from the road so they
would not be hit by a snowplow. On Nov 8, 2007 cable splicing was done in order to move the 0%
RCA multiplexer. However, due to snow, splicing was completed but the multiplexer was not
reinstalled and the other multiplexers were not moved at that time. The cables were spliced by
removing five millimeters to ten millimeters of the protective coating to expose each individual wire.
About 20 mm 25mm of the first shrink tubing was placed on each individual wire, and 50 mm
75mm of the second shrink tubing was placed over the cable. The exposed wires were twisted
together and flux was applied. The individual wires were soldered and the shrink tubing heated to
seal the cable. The splice was then wrapped with ducktape.
October 27 to 31, 2008, cable splicing for 30% and 50% sections was completed. All of the
multiplexers and dataloggers (0%, 30%, and 50% sections) were moved along the fence line and the
15% section was moved inside the traffic control cabinet with the datalogger. The solar panel for the
datalogger was also installed. All of the sensor cables joining the multiplexers to the datalogger were
relocated as well.
Figure 5.25 shows the relocated multiplexer and dataloger cabinets and their wiring.
72
Figure 5.25 Relocated Multiplexers and Datalogger, and Wiring
73
Chapter 6
Test Results
6.1 Thickness
The design thickness of the JPCP was 250 mm. Three methods were used to measure the actual
thickness of the JPCP test sections. First, the height of the JPCP surface above the asphalt stabilized
OGDL was measured eleven times during paving. Second, the sixteen extracted cores were
measured. Third, the MIT Scan-2 measures the distance to the centre of the dowel bar. This
measurement is then added to the known height of the dowel bar, 125 mm above the OGDL. The
MIT Scan-2 measure was performed 36 times. Figure 6.1 shows the thickness measurements made by
the three methods.
Figure 6.1 Thickness of JPCP

The results of the thickness testing showed that most of the pavement was thicker than designed.
Only 7 measurements were below the 250 mm design thickness. All three test methods produced
very similar averages and standard deviations, paving 272.3 mm + 19.7 mm, core 269.8 mm + 18.1
mm, and MIT Scan-2 271.4 mm + 18.1 mm.) The thickest parts of the test track occurred in the 15%
and 30% RCA sections. Several reasons for the variation in thickness include many start and stop
points, grade of the test sections and deformation of the asphalt stabilized OGDL.
6.2 MIT Scan-2
The MIT Scan-2 was designed for locating dowel bars inserted using a dowel bar inserter (DBI) but
can also be used for dowels placed in baskets if the bars are epoxy coated insulating them from the
basket and if the transport ties on the basket are cut [Yu 2005]. The MIT Scan-2 unit works by
emitting an electromagnetic pulse creating a magnetic field in the dowel bars that is detected and
74
measured. Figure 6.2 shows how the MIT Scan-2 is centred over the joint and pulled across the
pavement.
Figure 6.2 MIT Scan-2

An FHWA study shows the MIT Scan-2 is accurate to five millimeters for horizontal and vertical
misalignment, eight millimeters side shift and four millimeters depth with a 95% confidence [Yu
2005]. The MTO tolerance for dowel bar alignment is set at fifteen millimeters for horizontal and
vertical misalignment and 40 mm for side shift.
On June 18, 2007, Dufferin Construction performed an MIT Scan-2 to determine dowel bar
orientation and depth in the concrete. The testing was set up to measure both lanes in the same pass.
It took between 3.5 to 5 minutes to setup over the joint, scan, and move to the next joint. Figure 6.3
shows the distribution of dowel bar misalignment. Of the 928 dowel bars, 340 (37%) met MTO
standard with less than fifteen millimeters vertical and horizontal displacement. There are several
reasons for the large portion of dowel bars that did not meet MTO standard. During paving there was
significant deformation of the asphalt stabilized OGDL and it is possible that dowel baskets were not
fastened securely.
Figure 6.3 Distribution of Dowel Bar Misalignment
75
A rating system was developed by Yu [2005] to evaluate the quality of dowel bar alignment. This
system uses a joint score for joint-by-joint evaluation and a rolling average joint score for a series of
five joints. The joint score (JS) is calculated by multiplying the number of dowel bars in each
misalignment category by its weighting factor, adding all of the categories together and one to the
total. The calculation uses the greater of the horizontal or vertical misalignments. The weighting
factors for the misalignment types are given in Table 6.1.
Table 6.1 Joint Score Weighting Factors

Misalignment (mm) Weighting Factor
d <= 15 0
15 < d <= 20 2
20 < d <= 25 4
25 < d <= 38 5
d > 38 10
A joint score (JS) less than five has a very low risk of joint locking, between five and ten has a low
risk of joint locking, between ten and fifteen has a moderate risk of joint locking, and greater than
fifteen has a high risk of joint locking. Figure 6.4 shows the joint scores. The average joint score is
78 with a standard deviation of 52.2. Only one of the 36 joints would be classified as a low risk for
joint locking. The other 35 joints had a JS greater than fifteen indicating a high risk for joint locking.
Figure 6.4 Joint Scores

In isolation, a locked joint should not influence pavement performance, however, a problem occurs
if consecutive joints lock. The rolling average joint score is calculated by averaging five joint scores,
the two before the joint, the evaluated joint, and the two joints after. Each individual joint score used
in the calculation can have a maximum value of ten. A rolling average joint score of ten has a high
76
risk of developing distress because of two or more locked joints. Figure 6.5 presents the rolling
average joint scores.
Figure 6.5 Rolling Average Joint Scores

Since an average rolling joint score of ten represent the possibility of two or more locked joints
giving a high risk of developing distress.
6.3 Compressive Strength
Compressive strength testing was performed using 150 mm x 300 mm cylinders consolidated by
vibration. Compressive strength testing was done according to ASTM C 39 - Standard Test Method
for Compressive Strength of Cylindrical Concrete Specimens on days seven, fourteen, and twenty-
eight [ASTM 2004]. Cylinders were demolded after 24 hours and cured in the moisture room before
testing. A concrete specimen end grinder was used to prepare the samples prior to testing to ensure
the ends of the cylinders were perpendicular. After end grinding the cylinders were placed back in
the moisture room for half an hour to restore the moisture content of the cylinders lost during end
grinding before testing. Load was applied at a constant rate without shock. Figure 6.6 shows the
compressive strength development from the test track mixes.
A cone fracture occurred in 23 cylinders (63.9%), the remaining thirteen cylinders (36.1%) had
both a cone and shear failure. Compressive strength increased from 0% RCA to 30% RCA. All of the
RCA containing mixes reached the 28-day design strength, however, the control section containing no
RCA did not. The control section is projected to reach the 28-day design strength at 36 days. The
results show an RCA content of 30% can be used without negatively affecting the compressive
strength.
A two-sample t analysis was performed to identify if a difference exists between the mean
compressive strengths at 28-days for the preliminary and field mixes. A difference between the
mixes would suggest a change to the mix had occurred. Testing was conducted at a 95% confidence
interval. Results of the t procedure are summarized in Table 6.2.
77
Figure 6.6 Compressive Strength Results
Table 6.2 Results of t-Ttest

Mix df tcalculated tcritical
0% Coarse RCA 2 22.45 4.303
15% Coarse RCA 1 3.47 12.71
30% Coarse RCA 1 4.88 12.71
50% Coarse RCA 1 12.2 12.71
For 15%, 30%, and 50% Coarse RCA content tcritical is greater the tcalculated indicating that there is no
evidence of a difference between the preliminary and field mixes. However, for 0% Coarse RCA
there is a difference between the mixes based on the measured 28-day compressive strength values.
6.4 Flexural Strength
Flexural strength was tested with 150 mm X 150 mm X 600 mm beams consolidated by vibration.
Beams were tested on day seven, and twenty-eight according to the three-point bending ASTM C 78
Standard Test Method for Flexural Strength of Concrete using Simple Beam with Third-Point
Loading [ASTM 2004]. The span length of the testing equipment was 450 mm and the load was
applied at a constant rate without shock. Figure 6.7 shows the testing apparatus with a fractured
flexural beam.
78
Figure 6.7 Flexural Strength Testing
The following equation was used to calculate the modulus of rupture in the middle third of the span
length.
PL
R= (6.1)
bd 2
If a fracture was outside the middle third of the span length by a maximum of five percent, then the
following equation was used.
!
3Pa
R= (6.2)
bd 2
where: R = modulus of rupture, MPa
P = maximum applied load indicated by the testing machine, N
!
L = span length, mm
!
b = average width of the specimen at the fracture, mm
!
d = average depth of the specimen at the fracture, mm
!
a = average distance between the line of fracture and beam nearest support, mm
!
!
All twelve flexural beams tested failed within the middle third of the span length. Figure 6.8 shows
! the results of the flexural strength testing.
79
Figure 6.8 Flexural Strength Testing Results
Flexural strength development followed a similar trend as compressive strength. Flexural strength
increased up to 30% RCA. The greatest increase in strength development from day seven to day 28,
was the 30% RCA, while the 50% RCA had the smallest increase in strength. Flexural strength
results are higher than the CAC recommended value of eight percent to twelve percent of the
compressive strength or 0.6 to 0.8 square root of compressive strength [Kosmatka 2002]. Flexural
strength values ranged from 17.3% of the compressive strength for the 30% RCA to 25.0% of the
compressive strength for 0% RCA.
6.5 Maturity
Maturity testing was performed according to ASTMC 1074 Standard Test Method for Estimating
Concrete Strength by the Maturity Method [ASTM 2004]. COMMAND Center maturity system
(sensors and software) by The Transtec Group was used to take time based temperature readings to
calculate maturity.
In order to create the maturity curve, two concrete cylinders were installed with COMMAND
Center sensors during the casting of the field specimens. The sensors were installed in the centre of
the cylinder and at mid-depth. Once the cylinders were demolded after 24 hours they were moisture
cured along with the other cylinders. Compressive strength testing was performed at seven, fourteen,
and 28-days. At the end of 28-days, the maturity curves were created for each RCA amount by
plotting a best fit line through the strength and maturity data.
To estimating in-place strength, two maturity sensors were attached to the strain gauge chair
structure during the placement of the test sections. Maturity sensors took readings every 20 minutes
for 28-days. This was used to estimate the concretes strength based upon the created maturity curves.
Maturity was calculated using the Nurse-Saul method (temperature-time factor). A semi-log maturity
curve was created. The sigmodial relationship developed for the HIPERPAV program could not be
used because of the upper predictive limit of 34.5 MPa (5,000 psi). The following equation was used
to calculate the maturity curves for each percent RCA that are shown in Figure 6.9.
80
M ( t ) = $ (Ta " To )#t (6.3)
where: M ( t ) = temperature-time factor at age t, degree-day or degree-hour

"t = time interval, days or hours
!
Ta = average concrete temperature during time "t , oC
!
To = datum temperature, o C
!
! !
!
Figure 6.9 RCA Maturity Curves

Estimates of in-place strength were developed using the maturity curves and are shown in Table
6.3.
Table 6.3 Strength Estimates of Placed Concrete

Time In-Place Compressive Strength (MPa)
0% RCA 15% RCA 30% RCA 50% RCA
1-day 10.9 16.2 13.9 11.0
3-day 16.5 24.7 24.0 19.5
7-day 20.9 31.3 31.3 25.9
14-day 24.3 36.6 37.4 31.0
28-day 28.0 42.1 43.5 36.2
The estimated strength of the placed concrete was the same order as the cylinders, 30% RCA had
the greatest strength and 0% RCA was the lowest. Strength developed faster in the field than the lab
due to higher air temperatures accelerating curing and causing a loss of moisture for hydration in the
concrete. Moisture was lost through evaporation, absorption into the coarse RCA, and being drawn
into the asphalt-stabilized OGDL. At seven days, the strength development of the placed concrete
was between 1.1 MPa (50% RCA) to 2.5 MPa (0% RCA) higher than the cylinders in the lab. At 28-
days, the placed concrete was 0.3 MPa (0% RCA) to 1.4 MPa (30% RCA) higher than the lab.
81
6.6 Freeze-Thaw
Freeze-thaw testing was performed according to the ASTM C 666 Procedure A Standard Test
Method for Resistance of Concrete to Rapid Freezing and Thawing and ASTM C 215 Standard Test
Method for Fundamental Transverse, Longitudinal, and Torsional Resonate Frequencies of Concrete
Specimens [ASTM 2004]. The optional length change test was not conducted. The samples were
demolded after 24 hours and cured for eleven days before being placed in saturated lime water at
room temperature for 48 before the start of the test. Figure 6.10 shows the beams placed in a freeze-
thaw bed that cycled from -20 o C to 4 oC and back every 220 minutes (3.67 hours).
Figure 6.10 Beams in the Freeze-Thaw Bed

The beams were completely submerged in water and rotated in orientation and position within the
freeze-thaw bed during testing to reduce effects from hotter of cooler areas.
The half of the freeze thaw bed not occupied by samples contained pervious concrete or dummy
specimens. An oscilloscope and Erudite MkIV by CNS Farnell were used for the testing the relative
dynamic modulus (RDM) of elasticity as shown in Figure 6.11. The frequency range used for all
tests was 1,000 Hz to 2,000 Hz. For each beam, four fundamental transverse frequency sweeps, one
sweep on each side, were conducted to calculate an average.
Figure 6.11 Erudite MkIV Testing Equipment and Setup
82
Testing was interrupted at cycle 214 for a period of two weeks to isolate and repair a loose
connection causing a short in the wires connecting the contact vibrator and the electronics unit.
The relative dynamic modulus of elasticity was calculated using the following equation.
" n2 %
Pc = $ 12 ' (100 (6.4)
#n &
where: Pc = relative dynamic modulus of elasticity, after c cycles of freezing and thawing, %
n = fundamental transverse frequency at 0 cycles of freezing and thawing
!
n1 = fundamental transverse frequency at c cycles of freezing and thawing
!
The durability factor was calculated using the equation below.
!
!
PN
DF = (6.5)
M
where: DF = durability factor of the test specimen
P = relative dynamic modulus of elasticity at N cycles, %
!
N = number of cycles at which P reaches the specified minimum value for discontinuing the
!
test or the specified number of cycles at which the exposure is terminated, which ever is less
!
M = specified number of cycles at which the exposure is to be terminated
!
Table 6.4 summarizes the properties of the beams before testing.

!
Table 6.4 Properties of Beams Prior to Testing
Beam Width Depth Length Fundamental Transverse Frequency (kHz) Mass
ID (mm) (mm) (mm) [1] [2] [3] [4] [Avg] (kg)
0% FT1 102 78 407 1.791 1.793 1.772 1.774 1.783 7.552
0% FT2 101 78 407 1.749 1.749 1.738 1.742 1.745 7.431
15% FT1 102 78 406 1.820 1.820 1.818 1.813 1.818 7.721
15% FT2 102 78 407 1.889 1.889 1.887 1.888 1.888 7.865
30% FT1 101 77 407 1.785 1.781 1.760 1.759 1.771 7.660
30% FT2 102 77 408 1.767 1.769 1.782 1.787 1.776 7.638
50% FT1 101 78 406 1.739 1.739 1.739 1.734 1.740 7.538
50% FT2 101 78 406 1.759 1.750 1.757 1.769 1.759 7.514
83
Table 6.5 shows the results at the end of the 300 freeze-thaw cycles for relative dynamic modulus
(RDM) and weight change. The RDM is similar for all of the beams, approximately 90%.
Table 6.5 Freeze-Thaw Testing Results

Beam RDM Mass
ID (%) (%)
0% FT1 91.2 -0.54
0% FT2 89.3 -0.54
15% FT1 90.3 -0.62
15% FT2 89.7 -0.54
30% FT1 92.0 -0.68
30% FT2 86.2 -0.84
50% FT1 88.9 -0.19
50% FT2 91.3 -0.35
The amount of mass loss increases as the RCA content increases to 30% and decreases at 50%.
Figure 6.12 shows the change in RDM of beams over the 300 cycles. There is a similar trend for all
RCA contents with the majority of the change in RDM occurring during the first 117 cycles. There
was not a continuous decrease in RDM values throughout the testing. Each RCA content had an
increasing trend for a segment of the freeze-thaw cycles.
Figure 6.12 Changes in Relative Dynamic Modulus
84
6.7 Coefficient of Thermal Expansion
AASHTO TP60 is the provisional standard method for measuring CTE. It is also the method
recommended for use in the MEPDG. Steps to measure CTE using AASHTO TP60 include
[AASHTO 2004]:
1. Soaking a 100 mm (four inch) or 150 mm (six inch) core in water for a minimum of 48 hours
before testing begins.
2. Measuring the length of the saturated core using calipers.
3. The core is placed in a support frame and submerged in a water bath.
4. The temperature of the water bath is changed from ten degrees Celsius (50 oF) to 50 oC (120
o
F) where it remains until three consecutive Linear Variable Differential Transformer
(LVDT) readings, taken 10 minutes apart, change by less than 0.00025 mm (0.00001 in.).
Then this step is repeated several times.
Since the AASHTO TP60 requires specialized equipment that is often not readily available, a
simplified approach was used to calculate CTE. The method used to evaluate CTE is adapted from
ASTM C 531, Standard Test Method for Linear Shrinkage and Coefficient of Thermal Expansion of
Chemical-Resistant Mortars, Grouts, Monolithic Surfaces, and Polymer Concretes [West 2005,
ASTM 2004]. The following are the steps to perform this simplified method for CTE testing.
1. Cores dry for a period of 28 days after being taken from the field.
2. Reference points are installed along the vertical axis of a cylinder 200 mm apart and allowed
to cure for five hours.
3. The air temperature is cycled between 20 oC (68 oF) and -20 oC (-4 oF). It remains constant at
each of these temperatures for a minimum of sixteen hours to ensure that the cores reach the
ambient temperature. Mechanical strain readings where taken at each of the temperatures. In
this study the temperature was held constant for 24 hours and two cycles were completed.
The temperature was controlled using a large walk-in freezer.
Both the AASHTO TP60 and the modified method calculate strain using equation 7.6 [West 2005,
AASHTO 2004].
#L
" PCC = (6.6)
(#T $ L)
where: " PCC = coefficient of thermal expansion (CTE), strain per o C

!
"L = change in unit length due to a temperature change "T
"T = change in temperature (T2 T1), oC
!
L = length of specimen, mm
! !
! 85
!
Using the measured lengths, strain and the corresponding CTE values were calculated. These
values are presented in Table 6.6. Strain was calculated by comparing the change of the measured
length with the original length.
Table 6.6 CTE Testing Results

0% RCA
Cycle A1-1 A1-2 A1-3 A2 A3 A4
1 2
L L L L L L
-56 6.90 -58 7.03 -59 7.34 -57 6.65 -65 7.89 -68 8.11
1
-10 5.88 2 7.49 -4 6.94 1 7.03 -5 7.66 3 8.97
-68 7.22 -54 6.96 -57 6.69 -64 8.04 -58 6.66 -46 6.29
2
-3 7.78 2 6.21 7 7.80 8 8.64 6 7.89 6 6.64
Average 6.95 6.29 7.19 7.59 7.53 7.50
15% RCA
Cycle B1 B2 B3 B4-1 B4-2 B4-3
-46 5.42 -49 5.86 -49 5.86 -56 7.06 -47 5.89 -48 6.02
1
1 5.92 -4 5.70 4 6.68 6 7.88 3 6.45 6 6.86
-58 7.7 -53 6.30 -54 7.46 -51 7.17 -55 7.34 -43 6.22
2
0 7.34 -1 6.39 5 7.43 4 6.75 6 7.56 -1 5.20
Average 6.60 6.06 6.86 7.22 6.81 6.08
30% RCA
Cycle C1 C2 C3-1 C3-2 C3-3 C4
-40 4.85 -33 3.97 -48 5.92 -39 4.88 -44 5.60 -39 4.72
1
-5 4.41 5 4.78 6 6.80 -1 4.88 1 5.80 6 5.61
-47 5.32 -36 5.10 -39 5.63 -47 5.84 -37 4.82 -40 5.76
2
-8 6.72 11 5.70 1 4.89 -14 4.09 5 5.17 2 5.13
Average 5.33 4.89 5.81 4.92 5.35 5.31
86
50% RCA
Cycle D1 D2-1 D2-2 D2-3 D3 D4
-33 4.03 -25 3.10 -34 4.25 -31 3.83 -33 4.09 -30 3.70
1
-5 3.50 5 3.71 -6 3.50 1 5.1 -3 3.76 -2 3.52
-35 4.74 -33 4.74 -32 3.26 -28 4.72 -36 4.07 -35 4.11
2
2 4.89 7 4.89 -1 3.81 8 4.41 3 4.70 9 5.31
Average 4.29 4.11 3.71 4.52 4.15 4.16
1. L (um), 2. (10-6 /oC)
The CTE values obtained for samples containing 15% RCA and 30% RCA are within the typical
range of values for a concrete made with a limestone aggregate (5.9x10-6 /oC to 9.2x10-6 /oC). Figure
6.13 shows that CTE decreased as the amount of coarse RCA increased. This relationship is strong as
demonstrated by the high R2 value. The decreasing CTE trends is due to replacing larger amount of
virgin limestone aggregate with the coarse RCA concrete that has a lower CTE value.
Figure 6.13 CTE Values for RCA

Higher CTE values are associated with increased joint deterioration, expand/shrink (slab curl), and
cracking, a lower CTE is more desirable. This testing clearly shows that using a quality RCA gives a
lower CTE than virgin aggregate. This is contradictory to previous research by Yang et al. [2003].
Since the research by Yang et al does not state any specific information about the quality of RCA
used, it is likely that the RCA was a lower quality than in this study. A high quality RCA of a
specific size was used in this study, and is likely the reason for the difference in results.
A recent CTE study by Tanesi et al using the AASHTO TP 60 method with automated equipment
produced results with an average variability of 0.7x10-6 /oC [Tanesi 2007]. Variability in the study
was defined as
"# PCC = # PCC (max) $ # PCC(min) (6.7)
87
!
where: "# pcc = coefficient of thermal expansion (CTE) variability, strain per oC
" PCC(max) = maximum CTE test value, strain per oC

" PCC(min) = minimum CTE test value, strain per oC
!
!
Using this definition for variability, the simplified method used in this study had approximately
! double the variability at 1.53x10-6 /oC. Although the variability is increased with the simplified
method, it is still a useful method for calculating CTE and had a very strong correlation with the
amount of RCA in the concrete.
6.8 Testing Summary
Table 6.7 compares the characteristics of concrete containing coarse RCA with existing literature.
The coarse RCA used in this research showed improvements in compressive strength, flexural
strength, and CTE, while no difference in freeze-thaw durability was observed. These results
contrary to the existing literature can be explained by the high quality RCA used in the study. The in-
service concrete had sufficient strength and durability showing no visible signs of deterioration and
when crushed all visible contaminates were removed.
Table 6.7 CTE Testing Results

Literature [ACPA 2009] UW RCA
Compressive Strength Up to 24% lower Up to 50% greater
Flexural Strength Up to 10% lower Up to 25% greater
Freeze-Thaw Depends on air void system Similar results
Coefficient of Thermal Expansion Up to 30% greater Up to 45% lower
88
Chapter 7
Field Performance
7.1 Method
The performance of the four pavement test sections was evaluated using visual surveys conducted
following the MTO Manual for Condition Rating of Rigid Pavements [Chong 1995]. The MTO
uses distress type, severity, and density with a measure of ride quality to calculate a pavement
condition index (PCI) value. Equation 8.1 presents the MTO calculation for PCI.
(
PCI = 10 " 0.1" RCI " DMI " 0.924 + 8.856 ) (7.1)
where: PCI = Pavement Condition Index

RCI = Riding Comfort Index
!
DMI = Distress Manifestation Index, from Equation 8.2.
!
!
The DMI is a weighted total of all the observed pavement distress and is calculated using Equation
! 8.2.
% n (
' DMIMax # $W i ( si + di ) *
DMI = 10 " ' i=1 * (7.2)
' DMIMax *
' *
& )
where: DMI = Distress Manifestation Index
DMIMax = Theoretical maximum value, 196 for concrete
!
i = Distress type
!
W i = Pavement Distress Weight, values 0.5 to 3.0
!
si = Distress severity, values 0 to 4.0
!
di = Distress density, values 0 to 4.0
!
!
A summary of pavement distress types, weighting, and distress severity and density are presented
! in Tables 7.1 to 7.3.
89
Table 7.1 Pavement Distress Weights
Pavement Distress Weight (Wi)
Ravelling 0.5
Polishing 1.5
Scaling/Abrasion 1.5
Potholing 1.0
Joint Cracking/Spalling 2.0
Faulting 2.5
Distortion 1.0
Sealant Loss 0.5
Joint failure 3.0
Longitudinal Cracking 2.0
Transverse Cracking 2.0
Diagonal/Edge Cracking 2.5
Table 7.2 Distress Severity Levels and Weights

Severity Level None V. Slight Slight Moderate Severe V. Severe
Weight (si) 0 0.5 1 2 3 4
Table 7.3 Distress Density Levels and Weights

Density Level None Few Intermittent Frequent Extensive Throughout
Weight (di) 0 0.5 1 2 3 4
% Area 0 < 10 < 20 < 50 < 80 > 80
A description of the different pavement distress severities is presented in Appendix C. The PCI
value is used for determining maintenance options, pavement deterioration rates, and maintenance
priority scheduling; it is not a measure of structural capacity, surface roughness or skid resistance.
The PCI scale ranges from 100 (perfect/excellent) to zero (fully deteriorated/failed). Table 7.4
illustrates the PCI rating scale.
Table 7.4 PCI Rating Scale

Rating Excellent V. Good Good Fair Poor V. Poor Failed
PCI Value 85 - 100 70 - 85 55 - 70 40 - 55 25 - 40 10 - 25 0 - 10
90
7.2 Evaluations
Nine evaluations have been completed by the researcher and they were reviewed for consistency by
the supervisor who has experience in evaluations. The initial evaluation was conducted on June 18,
2007 five days after paving was completed. Additional evaluations were performed on September 20,
2007, November 9, 2007, April 4, 2008, June 25, 2008, September 21, 2008, November 1, 2008,
April 23, 2009, and June 6, 2009.
7.2.1 Evaluation #1 June 18, 2007
For this first evaluation, no deterioration was observed, and all test sections received a PCI of 100.
No shrinkage cracking was observed in any of the test sections. The pavement shoulders had not yet
been constructed. There was cutting residue along all joints, approximately 0.3 m to each side. The
entire pavement surface was covered in a fine dust. There was a rough pavement surface due to the
tining experiment. There were noticeable transverse dips in the pavement caused by a stationary
paver. Figure 7.1 show the general condition of the test section four days after construction. Figure
7.2 shows a close-up of some of the experimental tining
Figure 7.1 0% Coarse RCA Northbound View and 50% Coarse RCA Southbound View
Figure 7.2 Pavement Surface Tining Close-up
91
7.2.2 Evaluation #2 September 20, 2007
No change in performance was seen on the second evaluation. All of the test sections still have a PCI
of 100. The joints are widened and properly sealed. One joint in each of the four test sections NB
direction have small patches from checking the accuracy of the MIT-2 scan data. Gravel shoulders
are constructed.
7.2.3 Evaluation #3 November 9, 2007
The third evaluation was carried out approximately five months after the test sections were placed.
All of the test sections had some deterioration from the last evaluation but were all rated in excellent
condition with a PCI greater than 85. Table 7.5 summarizes the test track conditions observed during
the third evaluation.
Table 7.5 Evaluation 3 Test Track Conditions

Test Section PCI Pavement Distress Severity Density
0% Coarse RCA 97 Abrasion Very Slight Intermittent
Joint Cracking/Spalling Very Slight Intermittent
15% Coarse RCA 98 Joint Cracking/Spalling Very Slight Intermittent
The 0% coarse RCA test section is showing some minor amounts of abrasion loss. The exposed
interface at the South end of the control section experienced heavy loads from trucks turning onto the
test section from an unpaved extension. These trucks tracked gravel and debris onto the test section
that degraded the surface. Figure 7.3 shows the joint spalling that is common to all four test sections.
No cracking or deterioration of patches was noted. Figure 7.4 shows intermittent spalling of the
pavement edge and shoulder settlement of approximately 15% of all SB sections (loaded). Wide
loads not always traveling on the pavement surface cause this. The damage to the pavement edge is
not included in the PCI calculation because it does not directly affect the performance of the road.
Figure 7.3 Joint Spalling

92
Figure 7.4 Pavement Edge Spalling and Shoulder Settlement
7.2.4 Evaluation #4 April 4, 2008
Each test section had some additional deterioration compared to the last evaluation. All of test
sections are still in excellent condition with a PCI greater than 85. Table 7.6 summarizes the
conditions observed in the fourth evaluation. Figure 7.5 portrays some of the deterioration observed
during this evaluation. The abrasion of the 0% coarse RCA test section has been accelerated because
of some water pooling at the end of the end of the test section. Figure 7.6 shows a core hole with
settlement surrounding the hole but no cracking. Figure 7.7 shows the transverse depressions due to a
stationary slipform paver. The cracking of the asphalt adjoining the test sections is shown in Figure
7.8. Finally, Figure 7.9 shows the failure of the blockout containing inter-panel sensors spanning the
joint in 0% and 15% Coarse RCA. These areas will need repair.

0% Coarse RCA 92 Abrasion Severe Frequent
Joint Cracking/Spalling Moderate Frequent
15% Coarse RCA 96 Joint Cracking/Spalling Moderate Frequent
Ravelling Very Slight Few
93
Abrasion Joint Craking/Spalling
Ravelling Edge and Shoulder Settlement

Figure 7.5 Evaluation 4 - Pavement Distresses
Figure 7.6 Core Hole Settlement
94
Figure 7.7 Transverse Depression from Slipform Paver
Figure 7.8 Asphalt Adjoining Test Sections
Figure 7.9 Blockout Failure
95
The test sections remain unchanged from the last evaluation. Table 7.7 shows that all of the test
sections are in excellent condition with a PCI greater than 85.

Figure 7.10 shows the extensive spalling, approximately 80%, of the pavement edge and shoulder
settlement in all sections of the SB (loaded) direction. Settlement of the shoulder has increased to
over 25mm.
Figure 7.10 Pavement Edge Spalling and Shoulder Settlement
7.2.6 Evaluation #6 September 21, 2008
Compared to evaluation five, evaluation six showed some minor deterioration with very slight
raveling of three sections (0%, 15%, and 50%). All of the test sections are still in excellent condition
with a PCI greater than 85. Table 7.8 summarizes the conditions observed during evaluation six.
Figure 7.11 to Figure 7.13 show some of the pavement distress. Joint spalls with loose or missing
pieces are shown in Figure 7.11, new raveling is shown in Figure 7.12 and an increase in severity and
density of asphalt cracking at the start of the test sections is shown in Figure 7.13.
96
Figure 7.11 Joint Spalling
Figure 7.12 New Raveling

97
Figure 7.13 Asphalt Cracking
An automated distress survey was conducted of the test sections during evaluation six; however,
the results are currently unavailable.
7.2.7 Evaluation #7 Nov 1, 2008
The test sections remain unchanged since the last evaluation. All of the test sections are in excellent
condition with a PCI greater than 85. Table 7.9 summarizes the conditions of the test sections.

98
7.2.8 Evaluation #8 April 23, 2009
Abrasion is now evident in all test sections and is the worst in the 0% and 30% coarse RCA sections.
At these points large amounts of gravel, dirt, and garbage are being placed on the test sections and
driven over causing the distress. All of the test sections are still in excellent condition with a PCI
greater than 85. Table 7.10 outlines the observed condition of the test track.

0% Coarse RCA 91 Abrasion Severe Extensive
Abrasion Very Slight Few
Abrasion Slight Intermittent
Scaling in 0% coarse RCA test section has increased in density to almost 70% of the area. Figure
7.14 shows the spalling of the pavement edge and shoulder settlement in all sections. This spalling is
almost 100% of the SB (loaded) direction. The settlement of the shoulder has also increased to over
50 mm in some areas. The asphalt adjoining the test sections had additional deterioration over the
winter so some patching was attempted of the area as shown in Figure 7.15.
Figure 7.14 Pavement Edge Spalling and Shoulder Setlement
99
Figure 7.15 Patching of Adjoining Asphalt
There was only one change, a new pothole, from the previous evaluation. All test sections continue to
be in excellent condition with a PCI greater than 85. Table 7.11 and Figure 7.16 summarize the test
section conditions. Figure 7.17 shows the new pothole in the 0% coarse RCA test sections looks to
have been caused by machinery. The pavement shoulders are shown in Figure 7.18.

0% Coarse RCA 91 Abrasion Severe Extensive
Pothole Slight Few
Abrasion Slight Intermittent
100
0% Coarse RCA 15% Coarse RCA
30% Coarse RCA 50% Coarse RCA

Figure 7.16 Evaluation 9 Test Section Conditions
Figure 7.17 New Pothole in 0% Coarse RCA
101
Figure 7.18 Regraded Southbound Shoulder
7.3 Summary
The results of the nine evaluations over the previous 24 are shown in Figure 7.19. All of the test
sections are in excellent conditions with a PCI value greater than 85 and are showing no significant
difference in performance. The PCI decreased the most after each winter season.
Figure 7.19 PCI Progression
102
Chapter 8
Sensor Evaluation
Each test section had one slab that was instrumented with twelve sensors. Each instrumented slab
contained six vibrating wire concrete embedment strain gages to measure long term longitudinal and
transverse strain due to environmental changes, two vibrating wire vertical extensometers to monitor
slab curling and warping, two vibrating wire inter-panel extensometers to monitor joint movement,
and two maturity meters to measure maturity and temperature.
On September 6, 2007 continuous readings started. On November 8, 2007 the 0% RCA
multiplexer was disconnected to lengthen the sensor cables and to relocate the multiplexer cabinet.
The waste management facility staff buried the cable before it was reconnected. A snowplow cut the
cable connecting the 30% and 50% RCA mutiplexers to the datalogger in December 2007 and the
15% RCA multiplexer cable was cut in February 2008.
On October 31, 2008 cable splicing was done to add 10m-13m of cable for each sensor in the 30%
and 50% RCA sections and relocation of the 4 remote multiplexer cabinets and datalogger were
completed. Continuous monitoring was restarted November 8, 2008. When the data was downloaded
on November 15, 2008, there was an error for every sensor at each time interval. The sensor errors at
each interval for every sensor were NAN (not-a-number) for strain and displacement measurements
and -104 for the temperature measurements. The NAN error can be due to the voltage exceeding the
specified range or an open circuit. The temperature is calculated using Equation 8.1.
Temperature = "104.78 + 378.11# " 611.59# 2 + 544.27# 3 " 240.91# 4 + 43.089# 5 (8.1)
where: " = sensor reading
!
Many steps were taken to solve the error readings. The system was checked after each step and
! since error messages were still showing then further steps were carried out.
1. The connections were checked and adjusted to ensure there was not an open circuit from
wires touching each other. The wiring was checked against the wiring diagram.
2. The order of wiring for the temperature side of the connection was switched.
3. The datalogger program was compared to the Interlocking Concrete Paver (ICP) Project
datalogger program and a number of differences were noted. The delay value was 5 for
RCA and 15 for ICP. The end frequency was 1,200 for RCA and 1,000 for ICP. Finally
the time sweep was 250 in the RCA program and 200 in the ICP program.
4. Several modifications were made to the computer program. The order that the remote
multiplexers were called was changed to one then two, instead of two then one. The
voltage range was adjusted (2.5, 7.5, 25, and 250 mV) for the measurement of strain and
103
displacement. The start and ending frequency were modified (20 to 5000 Hz) for the
measurement of strain and displacement. The voltage range (2.5, 7.5, 25, 250, 2500 mV,
and autorange) for the measurement of temperature was adjusted. Finally, the frequency
range on the strain and displacement measurement was increased.
5. RST Instruments was contacted for assistance and suggested that an open circuit was
causing the errors.
6. Again the wiring and connections were checked and adjusted. The protected line was
connected correctly, however, the in/out of the lighting protection board had two wires
connected incorrectly. Once these wires were fixed, the errors for multiplexers one and
two were resolved, however, multiplexer three and four still had the same error.
7. RST Instruments was again contacted and recommended that the delay be increased.
8. A new cable was installed to run from the datalogger to multiplexers three and four. The
length of the delay value was increased to account for the added wire length to each sensor.
This resolved the errors for multiplexers three and four.
On March 6, 2009 the NAN and temperature errors were resolved and continuous monitoring was
restarted.
8.1 Datalogger Program
RST Instruments developed the datalogger program used to collect the sensor data. This program
records static sensor readings every two hours. The remote multiplexers are turned on and off in
series, multiplexers one for 15% RCA and two for 0% RCA, then three for 30% RCA and four for
50% RCA. Each multiplexer takes the readings from six strain gauges first, followed by the two
vertical extensometers, and two inter-panel extensometers. Sensor readings are converted into strains
and displacement. James Smith and Terry Ridgeway made modifications to the program to match the
sensor serial numbers for the vertical and inter-panel extensometers to their proper channel and to
increase the amount of time between reading successive sensors to account for the longer cable
lengths. The revised datalogger program is presented in Appendix D.
8.2 Sensor Readings
8.2.1 Strain Gauges
The impact of weather on JPCP performance can be monitored through strain data. Strain is the ratio
of an objects change in length to its original length due to an external action (load, temperature,
etc.). Using strain information the onset of cracking can be identified and predicted. Crack formation
in a 30 MPa concrete in compression (crushing) and tension (flexural) is approximately 3500 e and
130 e respectively [Neville 1995].
To calculate the strain caused by changes in climate, the raw strain data recorded by the datalogger,
Figure 8.1, is zeroed and corrections made to account for temperature variations and coefficients of
104
thermal expansion based on Equation 8.2. A positive strain value denotes that the pavement is in
tension, and a negative strain value shows compression.
March April May June
Figure 8.1 Sample Raw Strain Data, 15% Coarse RCA
$F'
"(Calculated ) = ( R1 # R0 )& ) + (T1 # T0 )(C1 # C2 ) (8.2)
% F0 (
where: R0 = initial strain reading
R1 = subsequent strain reading
!
F = strain gauge calibration factor, 3.405
!
F0 = default strain gauge factor, 4.062
!
T1 = initial temperature reading (oC)
!
T0 = subsequent temperature reading (oC)
!
C1 = coefficient of expansion of gauge (12.2 " )
!
C2 = coefficient of expansion of concrete ( " )
!
! !
Figure 8.2 presents the calculated strain profiles for four sensors in the 15% coarse RCA test
! section over a four-month period from !March to June 2009. The sensors include: S2 transverse
orientation located centre slab in the upper part of the JPCP; S1 transverse orientation located centre
slab in the lower part of the JPCP; S3 longitudinal orientation located centre slab in the lower part of
the JPCP; and S6 transverse orientation located at the join in the upper part of the slab. Table 8.1
highlights the minimum and maximum strain values for the sensors.
105
Figure 8.2 Calculated Strain Profile, 15% Coarse RCA
Table 8.1 Maximum and Minimum Strain Values

Transverse Transverse Longitudinal Joint
Upper Lower Lower Upper
(S2) (S1) (S3) (S6)
Compression strain
Calculated strain value () -79.1 -53.7 -61.9 -79.7
Date Mar. 13 Mar. 13 Mar. 13 Mar. 12
Time 08:00 08:00 08:00 08:00
Concrete temperature (oC) -7.4 -5.1 -5.2 -6.4
o
Ambient temperature ( C) -8.4 -8.4 -8.4 -8.3
Tension strain
Calculated strain value () 64.4 84.6 62.6 69.3
Date Apr. 17 Jun. 25 Apr. 18 Jun. 16
Time 16:00 06:00 16:00 16:00
Concrete temperature (oC) 23.3 28.5 19.3 29.2
o
Ambient temperature ( C) 18.9 20.8 20.2 23.6
Change in calculated strain () 143.5 138.3 124.4 149.0
o
Change in concrete temperature ( C) 30.7 33.6 24.5 35.6
Change in ambient temperature (oC) 27.3 29.2 28.6 31.9
106
Several observations were made from the strain results.
The maximum calculated strain values are between 1.5% and 2.5% required for crushing, and
between 48% and 65% necessary for cracking.
The change in strain is greater for the sensors located in the top portion of the slab compared
to the bottom. This is due to relative changes in temperature of each senor. The upper
portion of the slab has less cover and would be affected by constant changes in ambient
temperature. The lower portion of the slab has more cover and is not influenced as much by
the temperature changes. Therefore the greater the range of temperature the sensor
experiences the greater the induced strain.
The change in strain is greater for the sensors placed in the transverse direction compared to
the longitudinal. The difference in the strain results is due to the difference in the number of
restrained boundaries of each direction. Sensors in the transverse have only one restrained
edge at the centre of the pavement allowing for greater slab movement for a given
temperature. The sensors longitudinal are restrained by both joints at either end of the
pavement reducing the slab movement in that direction.
The change in strain is constant for the sensors located in the transverse direction located at
the mid slab and at the joint. Strain at mid slab and at the joint should be equal when placed
at the same depth. This is because the movement is restrained by only the pavement
centreline and consistent temperature.
Additional strain results for 0%, 15%, 30%, and 50% Coarse RCA are presented in Appendix E.
8.2.2 Vertical Extensometers
Slab curling is caused by temperature gradients that develop when one side of the pavement slab is
either warmer or cooler than the other. Figure 8.3 illustrates the curling of the pavement during the
day and night [MDOT 2007].
Figure 8.3 Slab Curling
During the day the surface of the slab is warmer than the bottom of the slab causing a positive
temperature differential (warmer on top and cooler on the bottom) creating a convex shape where the
107
middle of the slab is higher than the edges. In this position the surface is in tension and the bottom is
in compression. The opposite is observed during the night when a negative temp gradient is created
resulting in a concave shape of the slab with the edges being higher than the middle of the slab.
Figure 8.4 shows sample vertical displacement profiles at the edge of pavement over a four-month
period from March to June 2009 for the 15% coarse RCA sensors. The normalized vertical
displacements for the edge of pavement and centre slab presented in Figure 8.5.
Figure 8.4 Sample Vertical Displacement, 15% Coarse RCA
Figure 8.5 Normalized Vertical Displacement, 15% Coarse RCA
Several observations can be noted from the vertical displacement. The shape of the pavement slab
becomes more convex as the ambient temperature increases at the edge of the pavement. The
opposite trend is true for the vertical extensometer readings at the centreline. There is a greater
108
amount of displacement at the edge of the pavement. The temperature recorded by the sensor follows
the same trend as the ambient temperature but with a time lag.
Table 8.2 shows the minimum and maximum vertical displacement values for sensors S1 and S2.
Table 8.2 Maximum and Minimum Vertical Displacement Values

Edge Centreline
(S1) (S2)
Maximum displacement
Vertical displacement (mm) -4.52 -6.25
Date Jun. 23 Mar. 14
Time 20:00 10:00
o
Concrete temperature ( C) 32.10 -2.30
Ambient temperature (oC) 24.79 1.42
Minimum displacement
Vertical displacement (mm) -3.66 -5.88
Date Mar. 13 Jun. 25
Time 08:00 06:00
Concrete temperature (oC) -4.30 28.58
o
Ambient temperature ( C) -8.43 20.83
Change in displacement (mm) 0.86 0.27
o
Change in concrete temperature ( C) 36.4 30.88
Change in ambient temperature (oC) 33.22 19.41

The change in vertical displacement is three times greater at the edge of pavement compared
to the centreline.
Additional vertical displacement results for 0%, 30%, and 50% Coarse RCA are presented in
Appendix E.
8.2.3 Inter-Panel Extensometers

The change in temperature throughout the day causes the pavement slab to expand and contract.
During the daytime when the surface temperature is higher than the bottom of the slab, the slab
expands, reducing the space between the joints. Figure 8.6 shows sample horizontal displacement
109
profiles at the edge of pavement over a four-month period from March to June 2009 for the 15%
coarse RCA sensors. The normalized horizontal displacements for the edge of pavement and centre
slab presented in Figure 8.7.
Figure 8.6 Sample Horizontal Displacement, 15% Coarse RCA
Figure 8.7 Normalized Horizontal Displacement, 15% Coarse RCA
The data shows that the space between adjoining slabs decreases as the ambient temperature
increases. There is a greater amount of displacement at the edge of pavement. Joint movement
follows the same trend as the ambient temperature, but with a time lag. Table 8.3 shows the
minimum and maximum horizontal displacement values for sensors S5 and S6.
110
Table 8.3 Maximum and Minimum Horizontal Displacement Values
Centreline Edge
(S1) (S2)
Maximum displacement
Horizontal displacement (mm) -6.20 -4.42
Date Mar. 20 Mar. 14
Time 10:00 14:00
Concrete temperature (oC) -0.05 -1.00
o
Ambient temperature ( C) -2.18 5.36
Minimum displacement
Horizontal displacement (mm) -6.64 -4.96
Date Jun. 25 Jun. 24
Time 20:00 20:00
o
Concrete temperature ( C) 27.47 30.67
o
Ambient temperature ( C) 20.94 24.83
Change in displacement (mm) 0.44 0.54
Change in concrete temperature (oC) 27.52 31.67
o
Change in ambient temperature ( C) 23.12 19.47

The change in horizontal displacement is equal along the joint at the edge of pavement and
centerline locations.
Additional horizontal displacement results for 0%, 15%, and 50% Coarse RCA are presented in
Appendix E.
The movement of longitudinal joints in a JPCP is influenced by four key factors. Theses factor
include: temperature change, slab length, CTE, and frictional forces created between the base and
slab. Equation 8.3 presents the AASHTO 1986 Guide for the Design of Pavement Structures
formula to estimate joint movement [FHWA 1990].
z = CL(e"T ) (8.3)
where: z = joint opening (in)

C = base/slab frictional restraint factor (0.65 for stabilized bases, 0.8 for granular bases)
!
L = slab length (in)
! 111
!
!
e = PCC coefficient of thermal expansion ( /oF)
"T = maximum temperature range (oF)
!
To validate the inter-panel extensometer results, sensor readings were combined with the calculated
! CTE values of the RCA mixes. Values used in the calculation are shown in Table 8.4.
Table 8.4 Joint Movement Calculation Values

RCA Content 0% Coarse 15% Coarse 30% Coarse 50% Coarse
Sensor ID S3 S4 S1 S2 S5 S6 S7 S8
Slab length (m) 4.3 4.3 4.3 4.3
-6 o
CTE Maximum (10 / C) 8.97 6.86 5.76 5.31
CTE Average (10-6/oC) 7.28 6.60 5.27 4.10
CTE Minimum (10-6/oC) 6.29 5.20 4.72 3.52
o
Temperature change ( C) NA* NA* 27.52 29.94 39.03 30.25 28.32 24.51
* Inter-panel extensometers S3 and S4 are not working
The measured joint movement for the inter-panel extensometers falls within the calculated limited
of Equation 9.3 for all of the sensors S6 as illustrated by Figure 8.8. This confirms that the sensors
are working properly and the estimated CTE range of the RCA concretes.
Figure 8.8 Calculated and Measured Joint Movement
112
8.3 Predictive Models
Predictive models were created to estimate the temperature and strain or displacement readings of the
sensors. Predictive models for strain and displacement were created in a three-step process.
Step 1: Estimate the temperature for the strain and displacement sensors.
Step 2: Create predictive models for strain and displacement using the actual measured
temperature sensor data.
Step 3: Use the predictive models from Step Two with the temperature estimates from Step
One to predict raw strain and displacement values.
8.3.1 Temperature
In order to estimate the sensor temperature reading an average of the current and preceding ambient
temperatures was used. The same model was used for all RCA amounts.
Two temperature models were created for the strain sensors because they are located at different
depths in the pavement. Strain sensors located in the upper part of the pavement used an average of
the current and one preceding ambient temperature reading. Whereas strain sensors located in the
lower part of the pavement used an average of the current and five preceding ambient temperature
readings.
One temperature model was created for the vertical extensometers. The temperature was estimated
using an average of the current and six preceding ambient temperature readings.
One temperature model was created for the inter-panel extensometers. The temperature was
estimated using an average of the current and five preceding ambient temperature readings.
Figure 8.9 shows the correlation between the weighted ambient temperature and the measured
concrete temperature for the 15% RCA strain sensor located in the lower part of the pavement. The
R2 value is 0.847 showing good correlation between the predictive model and the actual sensor
measurements.
Figure 8.9 Strain Temperature Prediction

113
Table 8.5 shows the predictive ability of the temperature models based on the coefficient of
determination (R2) values.
Table 8.5 Temperature Coefficient of Determination Values

Strain
Transverse (Upper) 0.854 0.852 0.588 0.837
Transverse (Lower) 0.837 0.847 0.842 0.822
Longitudinal (Upper) 0.855 0.853 0.636 0.838
Longitudinal (Lower) 0.836 0.846 0.803 0.818
Joint (Upper) - 0.849 0.853 0.839
Joint (Lower) 0.780 0.847 0.830 0.819
Vertical Extensometer
Edge of pavement 0.764 0.881 0.878 0.850
Centreline - 0.879 0.758 0.868
Inter-panel Extensometer
Edge of pavement - 0.870 0.761 0.861
Centreline - 0.865 0.879 0.864
The models are able to predict the correct temperature within plus minus five degree Celsius. The
2
R is greater than 0.8 for 32 of 40 sensors demonstrating a high correlation. Four of the 0% RCA
sensors are yielding values that are more positive and negative than the corresponding ambient
temperatures indicating a failure of the sensor.
8.3.2 Strain
A model for each of the 24 sensors was created to predict the raw strain values recorded by the
datalogger. The strain for each of the coarse RCA contents and locations within the slab were
best described using a second-degree polynomial equation. Figures 8.10 and 8.11 show a sample
of the predictive ability of the strain equations for the 15% coarse RCA transverse orientation in
the upper and lower part of the slab.
114
Figure 8.10 Actual vs. Predicted Raw Transverse Strain, Centre Slab, Upper
Figure 8.11 Actual vs. Predicted Raw Transverse Strain, Centre Slab, Lower
The models for sensors in the lower part of the slab had a better predictive ability than those in the
upper part of the slab. The difference between the actual and predicted strain increase as the ambient
temperature increases with a greater discrepancy for the upper sensors. Table 8.6 presents the
predictive strain equations for each sensor.
115
Table 8.6 Predictive Strain Equations
Sensor Predictive Raw Strain Equation R2
ID Location
0% Coarse RCA
S7 Longitudinal (Raw) = -0.0254T2 2.7993T + 2671.6 0.78
S9 Transverse (Raw) = 0.0086T2 3.3884T + 2851.5 0.92
S10 Transverse (Raw) = -0.0615T2 0.0098T + 2555 0.91
S11 Joint NA
S12 Joint NA
15% Coarse RCA
S1 Transverse (Raw) = 0.0009T2 - 1.2727T + 3251.5 0.80
2
S2 Transverse (Raw) = 0.0149T 2.4403T + 2923.5 0.79
2
S3 Longitudinal (Raw) = 0.0198T - 1.2711T + 2910.6 0.61
2
S4 Longitudinal (Raw) = -0.0521T - 1.1931T + 3169.7 0.90
S5 Joint (Raw) = -0.0656T2 + 0.3693T + 2894 0.86
S6 Joint (Raw) = 0.013T2 2.3861T + 2142.4 0.81
30% Coarse RCA
S13 Transverse (Raw) = -0.0029T2 0.696T + 2062.6 0.81
S14 Longitudinal (Raw) = 0.1321T2 6.5697T + 15826 0.80
S15 Transverse NA
S17 Joint (Raw) = 0.0268T2 4.6955T + 14333 0.93
S18 Joint (Raw) = 0.0347T2 3.5693T + 2887.6 0.84
50% Coarse RCA
S20 Transverse (Raw) = -0.0201T2 1.3888T + 2973.8 0.89
S21 Transverse (Raw) = 0.0237T2 2.5257T + 3096.1 0.84
S23 Joint (Raw) = -0.0424T2 + 0.1077T + 2978.5 0.76
S24 Joint (Raw) = -0.0212T2 0.1235T + 2756.8 0.70
116
8.3.3 Vertical Extensometer
A model for each sensor was created to predict the vertical displacement. The displacement for each
of the coarse RCA contents and locations within the slab were best described using a second-degree
polynomial equation. Figures 8.12 and 8.13 show a sample of the predictive ability of the
displacement equations for 15% coarse RCA located at the edge of the pavement and at the
centreline.
Figure 8.12 Actual vs. Predicted Vertical Displacement at Edge of Pavement
Figure 8.13 Actual vs. Predicted Vertical Displacement at Centreline
Several observations were made about the predictive equations. First, they tend to over predict
vertical displacement at the edge of the pavement and under predict the vertical displacement.
117
Second, there was a consistent difference between the actual and predicted strain increase as the
ambient temperature increased. Table 8.7 presents the predictive vertical displacement equations.
Table 8.7 Predictive Vertical Displacement Equations

Sensor Predictive Vertical Displacement Equation R2
ID Location
15% Coarse RCA
S1 Edge D(Vertical) = 0.0002T2 0.0264T -3.8137 0.95
S2 Centreline D(Vertical) = 0.00003T2 0.0059T - 6.18 0.90
50% Coarse RCA
S7 Centreline D(Vertical) = -0.00001T2 0.002T - 5.4392 0.96
S8 Edge D(Vertical) = -0.0002T2 + 0.0014T - 5.6976 0.92
8.3.4 Inter-Panel Extensometer
A model was developed for each of the eight sensors to predict the horizontal displacement. The
displacement for each of the coarse RCA contents and locations within the slab were best described
using a linear equation. Figures 8.14 and 8.15 show a sample of the predictive ability of the
displacement equations for 15% coarse RCA located at the edge of the pavement and at the
centreline.
Figure 8.14 Actual vs. Predicted Horizontal Displacement at Edge of Pavement
118
Figure 8.15 Actual vs. Predicted Horizontal Displacement at Centreline
Comparing the actual and the predicted displacement results shows that the predictive equations
under estimate the horizontal displacement at both the edge of the pavement and at the centreline.
There is a consistent difference between the actual and predicted strain increase as the ambient
temperature increases. The predictive vertical displacement equations are shown in Table 8.8
Table 8.8 Predictive Horizontal Displacement Equations

Sensor Predictive Horizontal Displacement Equation R2
ID Location
15% Coarse RCA
S1 Centreline D(Horizontal) = 0.0112T 6.2798 0.88
S2 Edge D(Horizontal) = 0.0127T 4.5546 0.85
30% Coarse RCA
50% Coarse RCA
119
Chapter 9
Mechanistic-Empirical Pavement Design Guide Performance
9.1 ME-PDG Background and How It Works
The development of ME-PDG began in 1997 by the American Association of State and
Transportation Officials (AASHTO) Joint Task Force on Pavements (JTFP) under National
Cooperative Highway Research Program Projects 1-37 and 1-37a [ARA 2004]. The ability to
develop the ME-PDG came through advancements in computers, modeling technologies, vast
amounts of pavement performance data available though the SHRP and LTPP programs and more
rigorous pavement design procedures [Dore 2005].
A new pavement design guide was needed to more accurately simulate climate, pavement design,
changes in traffic and new materials. A design guide that could accurately model the conditions of a
specific location was needed. The new guide should provide multiple design options with longer
analysis periods. Truck volume and loading had increased significantly from the original American
Association of State Highway Officials (AASHO) Road Test. Changes in the truck characteristics
axle configurations and tire pressure also had to be accounted for. New asphalt and concrete
materials had been developed since AASHO and the ability to better characterize all pavement
properties was needed. The new pavement design guide would allow specific confidence levels as
opposed to applying factors of safety multipliers based on traffic levels alone.
The ME-PDG goes beyond designing for pavement thickness as in previous design guides. It looks
at the interaction of climate, materials, and traffic on the pavement structure using both mechanistic
and empirical model to calculate damage over time leading to the distresses observed. Mechanistic
results (stress and strains) are used to estimate pavement distress, where as empirical results are used
when data is calibrated to observed performance [Hall 2005]. Figure 9.1 illustrates this concept.
Figure 9.1 ME-PDG Flow Chart

The ME-PDG simulates both new and rehabilitation designs at three input levels. Level One
obtains the highest degree of accuracy by using specific site and material information. Level Two
120
uses values similar to the AASHTO Design Guides. Level Three uses the national average default
values based on data taken from LTPP in 2000. Regardless of the level used, the analysis procedure
is the same.
The ME-PDG requires a large amount of inputs, these include:
General site and project information (e.g., design type, design life, construction and
opening dates);
Analysis parameters (e.g., initial and terminal IRI, various forms of cracking, permanent
deformation), critical values, and reliability (used to account for error in predictions of
various distresses);
Traffic assumptions (e.g., lane alignment, baseline volume and future growth and
distribution of truck traffic);
Pavement structure (e.g., specification of layer thicknesses and material properties); and
Climate (e.g., station location and elevation, groundwater table level, hourly temperature
data).
The JPCP module of the MEPDG predicts pavement performance based on three parameters:
transverse cracking, joint faulting, and pavement roughness. The transverse cracking model,
Equations 9.1 and 9.2, calculates the percentage of slabs that fall within the specified project limits.
Both modes of transverse cracking top-down and bottom up are included in the model.
1
CRK = (9.1)
1+ FD"1.68
n i, j,k,l,m,n
FD = " (9.2)
N i, j,k,l,m,n
!
where: CRK = predicted amount of top-down or bottom-up cracking (fraction)
FD = total fatigue damage (top-down or bottom-up cracking)
!
n i, j,k, ... = applied number of axle load applications at condition i, j, k, l, m, n
!
N i, j,k, ... = allowable number of axle load applications at condition i, j, k, l, m, n
!
i = age
!
j = month
!
k = axle type
!
l = load level
!
m = temperature difference
!
n = traffic path
!
The faulting model, Equations 9.3 to 9.6, predicts the cumulative transverse joint faulting each month.
!
! 121
m
Fault m = # "Fault i (9.3)
i=1
( (
"Fault i = 0.001725 + 0.0008( FR)
0.25
))(FAULTMAX i#1
2
# Fault i#1 ) DE i (9.4)
! #m &# 0.4 &

FAULTMAX i = FAULTMAX 0 + 1.2%% " DE j ((%log 1+ 250(5.0)
$ j=1 '$
EROD
( ) (
'
(9.5)
!
(
FAULTMAX 0 = 1.29 + 1.1( FR)( 0.25
))
! $ $ P200 (WetDays) ''
0.4 (9.6)
%
( EROD
)
("curling )&log 1+ 250(5.0) # log&% PS
))
((
where: Fault m = mean joint faulting at the end of month m (in)

!
"Fault i = incremental monthly change in mean transverse joint faulting during month i (in)
FAULTMAX i = maximum mean transverse joint faulting for month i (in)
!
FAULTMAX 0 = initial maximum mean transverse joint faulting (in)
!
EROD = base/subbase erodibility factor
!
DE i = differential deformation energy accumulated during month i
!
"curling = maximum mean monthly slab corner upward deflection PCC
!
PS = overburden on subgrade (lb)
!
P200 = percent of subgrade material passing #200 sieve
!
WetDays = average annual number of wet days greater than 0.1 in rainfall
!
FR = base freezing index
!
!
The IRI model predicts the pavement roughness combining the results from the transverse cracking
! and faulting models with joint spalling and a site factor as per Equations 9.7 to 9.10.
IRI = IRI1 + 0.8203(CRK ) + 0.4417( SPALL) + 1.4929(TFAULT ) + 25.24 ( SF ) (9.7)
AGE # 100 &

SPALL = % ( "12( AGE ) +SCF ) (
(9.8)
AGE + 0.01 $1+ 1.005 '
!
122
!
SCF = "1400 + 350( AIR%)(0.5 + PREFORM ) + 3.4 (0.4 f 'c )
(9.9)
"0.2( FTCYC ( AGE )) + 43hPCC " 536WC _ Ratio
SF = AGE (1+ 0.5556FI )(1+ P200 ) "10#6 (9.10)
!
where: IRI = predicted IRI (in/mi)
!
IRI1 = initial pavement smoothness, measured as IRI (in/mi)
CRK = percent slabs cracked with transverse cracks (all severities)
!
SPALL = percentage of joints with spalling (medium and high severities)
!
TFAULT = total joint faulting cumulated per mi (in)
!
SF = site factor
!
SPALL = percentage joints spalled (medium- and high-severities)
!
AGE = pavement age since construction (years)
!
SCF = spalling prediction scaling factor
!
AIR% = PCC air content (percent)
!
PREFORM = 1 if perform sealant is present; 0 if not
!
f 'c = PCC compressive strength (psi)
!
FTCYC = average annual number of freeze-thaw cycles
!
hPCC = PCC slab thickness (in)
!
WC _ Ratio = PCC water/cement ratio
!
FI = freezing index (oF-days)
!
P200 = percent subgrade material passing #200 sieve
!
!
! 9.2 ME-PDG Sensitivity
NCHRP 01-47 Sensitivity Evaluation of ME-PDG Performance Prediction call for proposals
identified that the current sensitivity analysis being conducted do not consider the effect of varying
two or more input parameters in a systematic fashion. The University of Arkansas conducted one of
the most recognized sensitivity analysis in 2005 [Hall 2005]. In this study 29 variables within the
JPCP module were analyzed individually for their influence on the three performance models
123
(cracking, faulting, and roughness) in the ME-PDG. The results of the sensitivity analysis are
presented in Table 9.1.
Table 9.1 MEPDG Sensitivity Analysis

Input Parameter Performance Model
Cracking Faulting Roughness
Curl/Warp Effective Temperature Difference X X X
Joint Spacing X X X
Sealant Type
Dowel Diameter X X
Dowel Spacing
Edge Support X X X
PCC-Base Interface
Erodibility Index
Surface Shortwave Absorptivity X
Infiltration of Surface Water
Drainage Path Length
Pavement Cross Slope
PCC Layer Thickness X X X
Unit Weigh X X X
Poissons Ratio X
Coefficient of Thermal Expansion X X X
Thermal Conductivity X
Heat Capacity
Cement Type
Cementitous Content
Water-Cement Ratio
Aggregate Type
PCC Set Temperature
Ultimate Shrinkage at 40% Relative Humidity
Reversible Shrinkage
Time to Develop 50% of Ultimate Shrinkage
124
Input Parameter Performance Model
Cracking Faulting Roughness
Curing Method
28-day PCC Modulus of Rupture X X
28-day PCC Compressive Strength X X
The sensitivity analysis shows that the key input variables are curl/warp effective temperature
difference, joint spacing, dowel diameter, edge support, surface shortwave absorption, PCC layer
thickness, unit weight, Poissons ratio, coefficient of thermal expansion, thermal conductivity, 28-day
modulus of rupture, and 28-day compressive strength.
9.2.1 Experimental Design
To addresses the issues raised by NCHRP 01-07 a screening design was used to determine which
input parameters are the most influential prior to modeling the RCA concrete. Sensitivity analysis
was performed using ME-PDG Version 1.0 release May 24, 2007. A Plackett-Burman (PB) design
was selected for the sensitivity analysis because it is able to screen a large amount of variables using a
limited number of runs compared to a full- or partial-factorial design. The standard PB design is a
resolution III design. This means that the effects of main variables can be confounded with second
order interactions produce some inconclusive results by not being able to distinguish between the
main and second order interaction effects. Applying a foldover technique to the PD design allows
the resolution of the design to be increased to IV, confounding results between the main and third
order interactions. The foldover creates a mirror image of the design with +s becoming -s and
vice-versa and adds an extra factor to the design. Two separate seven factor PD designs with
foldover, eight factors total, were used to test the sensitivity of the input parameters in the JPCP
layer module and the design features module of the ME-PDG. Adding an additional run using the
midpoints (0) of the +s and -s was run for comparison of the results. If multiple runs of the
midpoints were conducted an error estimate could be calculated to be included in ANOVA. This
however, is not possible because the ME-PDG keeps returning the same value resulting in no error.
Tables 9.2 and 9.3 present the eight factors and values used to examine the sensitivity of the JPCP
layer and design feature inputs. A broad range of values was analyzed for each factor to determine
the sensitivity.
125
Table 9.2 JPCP Layer Sensitivity Factors and Values
JPCP Factors - + 0
A Unit Weight (kg/m3) 2240 2560 2400
B Poissons Ratio 0.15 0.25 0.2
o -6
C Coefficient of Thermal Expansion (per C x10 ) 6.0 9.2 7.6
o 1
D Thermal Conductivity (J/m-s- C) 1.5 2.4 1.95
E Heat Capacity (J/kg-oC) 630 1890 1260
F Cementitous Content (kg/m3) 300 360 330
G Water-Cement Ratio 0.33 0.53 0.43
2
H 28-day Compressive Strength (MPa) 25 45 35
1. A typical thermal conductivity values for a concrete made with limestone aggregate 3.2 J/m-s-o C [Neville
1995], however this value was reduced because of an ICM satiability error associated with this input parameter.
2. Alternatively 28-day modulus of rupture values could have been used to quantify strength development.
Table 9.3 Design Sensitivity Factors and Values

Design Factors - + 0
A Joint Spacing (m) 3.7 4.5 4.1
B Sealant Type Liquid Preformed Silicon
C Dowel Diameter (mm) 25.4 44.5 32
D Dowel Spacing (mm) 254 356 305
1
E Edge Support None Tied Shoulder Widened Slab
2
F PCC-Base Interface (months) 0 480 240
G Erodibility Index3 5 1 3
H Surface Shortwave Absorpitivity 0.7 1 0.85
1. Tied shoulder has a 40% load transfer efficiency, and widened slab has a width of 12 feet.
2. PCC-Base interface values represent the time of loss of friction.
3. Erodibility index values: 1 = extremely resistant, 3 = erosion resistant, and 5 = very erodeable.
To estimate the first and second order effects for each of the two sensitivity analyses seventeen
ME-PDG runs are required. Table 9.4 shows the generated design for each PB screening analysis.
Runs seven through sixteen represent the foldover that is used to create the extra variable H.
126
Table 9.4 PB Design Run Variability Levels
Run Variable
A B C D E F G H
1 - - - + + + - +
2 + - - - - + + +
3 - + - - + - + +
4 + + - + - - - +
5 - - + + - - + +
6 + - + - + - - +
7 - + + - - + - +
8 + + + + + + + +
9 + + + - - - + -
10 - + + + + - - -
11 + - + + - + - -
12 - - + - + + + -
13 + + - - + + - -
14 - + - + - + + -
15 + - - + + - + -
16 - - - - - - - -
17 0 0 0 0 0 0 0 0
9.2.1.1 Calculations
The effect of each variable is calculated using Equation 9.11.
Effect = ( Average response "+") " ( Average response """) (9.11)
If two or more centre points (0) are used the experimental error can be calculated and used in an
ANOVA (analysis of variance) test to determine the significance of the variables. To complete the
ANOVA table
! a sum of squares (SS) and mean square (MS) values need to be calculated.
2
SS = R2 k"2 ( Effect ) (9.12)
where: R = the fraction or complete replication of the design

127
!
!
k = number of variables
SS(Effect)
! MS = (9.13)
df
where: df = degrees of freedom
9.2.1.2 Interpretation of the

! Results
! To identify the influential parameters the calculated effects are plotted against their expected
normal value. Significant variables do not fall along the straight line plotted thought the data. Also,
the magnitude of the calculated effect is also important. Figure 9.2 presents the results of a simplified
three factor PB design, to be used for illustration
Figure 9.2 Example Results
Based on this example the variables A, C, and their interaction AC are influential over the
range of limits tested. It is important to note that because the interaction AC was identified as a
significant variable, the effects of A or C can not be talked about independently.
9.2.2 Results
The PB results of the 50-year ME-PDG simulations for percent slabs cracked, joint faulting and
pavement roughness and time of failure (TOF) are given in Tables 9.5 and 9.6.
128
Table 9.5 JPCP Factor Results
JPCP Factor Cracking (% slabs) Faulting (mm) Roughness (m/km)
Run 50-year TOF 50-year TOF 50-year TOF
1 0.0 NR 1.17 NR 2.38 NR
2 1.0 NR 1.47 NR 2.60 NR
3 0.3 NR 2.16 NR 2.84 48
4 8.0 NR 1.60 NR 2.65 NR
5 47.7 13 4.67 22 4.37 20
6 30.8 36 2.46 NR 3.26 41
7 5.4 NR 4.39 25 3.62 30
8 48.9 28 2.92 NR 3.75 35
9 99.8 3 2.92 NR 4.62 16
10 65.7 25 3.53 38 4.35 29
11 98.7 5 2.49 NR 4.39 20
12 44.7 32 3.33 43 4.05 31
13 16.4 49 1.07 NR 2.81 49
14 94.2 4 2.36 NR 4.34 20
15 1.8 NR 0.81 NR 2.59 NR
16 30.5 37 1.42 NR 3.12 44
17 4.7 NR 2.84 NR 3.22 41
Table 9.6 Design Factor Results

Design Factor Cracking (% slabs) Faulting (mm) Roughness (m/km)
1 0.1 NR 4.67 25 4.04 27
2 26.4 42 6.58 15 4.61 25
3 0.2 NR 3.48 37 3.11 39
4 41.3 28 8.66 8 5.04 17
5 0.9 NR 1.68 NR 2.82 48
6 14.3 NR 2.90 NR 3.26 41
7 0.5 NR 2.41 NR 2.67 NR
129
Design Factor Cracking (% slabs) Faulting (mm) Roughness (m/km)
8 7.5 NR 2.01 NR 2.44 NR
9 3.3 NR 1.73 NR 2.29 NR
10 0.0 NR 1.45 NR 2.27 NR
11 1.2 NR 2.49 NR 2.95 46
12 0.0 NR 1.04 NR 2.55 NR
13 0.2 NR 6.76 14 3.89 26
14 0.0 NR 3.35 45 3.05 41
15 0.7 NR 5.13 25 3.37 35
16 0.1 NR 4.45 29 3.50 31
17 4.7 NR 2.84 NR 3.22 41
Using the results of Table 10.5 and 10.6, the effects of the eight JPCP and design inputs variables
and second order interactions were calculated. Figures 9.3 to 9.8 show the effect values plotted
against the expected normal values for the JPCP and Design variables. In each of the six plots only
the significant variables are identified.
Figure 9.3 JPCP Crack Effect Results
130
Figure 9.4 JPCP Faulting Effect Results
Figure 9.5 JPCP Pavement Roughness Effect Results
Figure 9.6 Design Crack Effect Results
131
Figure 9.7 Design Faulting Effect Results
Figure 9.8 Design Pavement Roughness Effect Results
PB testing identified four independent variables and one interaction as having a significant
influence on the results obtained from the ME-PDG simulations. These variables included:
CTE,
Unit weight,
Edge support,
Dowel diameter, and
Interaction between joint spacing and surface shortwave absorption.
The majority of the significant variables are known quantities and fixed values for the design,
however, special care needs to be taken in determining the CTE value.
132
9.3 Modeling Test Sections
Since CTE was identified as a sensitive variable impacting the results of the ME-PDG, three
simulations were run using each of the four RCA amounts used in the test section (0%, 15%, 30%,
and 50% coarse RCA) for a total of twelve ME-PDG simulations. The average CTE, minimum CTE,
and maximum CTE values were simulated for each RCA amount. The pavement structure consisted
of 250 mm JPCP of varying RCA amounts, 100 mm asphalt-stabilized ODGL, and 450 mm granular
base material on a clay subgrade.
9.3.1 Materials
Table 9.7 summarizes the inputs for the JPCP and asphalt stabilized base layers. The values are
based on the laboratory testing of the construction material and represent Level One and Level Two
inputs.
Table 9.7 ME-PDG Inputs for JPCP and Asphalt Layers

Concrete
Cement Content (kg/m3) 315
w/c Ratio 0.43
Unit Weight (kg/m3) 2350 2325 2323 2306
Compressive Strength (MPa)
7-day 19.8 29.8 29.1 23.4
14-day 23.9 35.6 35.9 29.7
28-day 27.7 40.9 42.1 35.0
90-day 35.2 45.4 46.8 38.9
20-year/28-day 1.32
CTE (10-6 /oC)
Average 7.280 6.603 5.265 4.103
Maximum 8.973 6.857 5.761 5.314
Minimum 6.287 5.203 4.717 3.519
Ultimate Shrinkage at 40% R.H. (10-6) 594 654 689 736
133
Asphalt Stabilized Base
Binder Grade PG 65-28
AC Content (%) 2
Air Voids (%) 20
3
Unit Weight (kg/m ) 2015
Poissons ratio 0.35
Thermal Conductivity (J/m-s-oC) 1.2
Heat Capacity (J/kg-oC) 966
Gradation
% Retained 3/4 7.4
% Retained 3/8 72.5
% Retained #4 95.9
% Passing #200 1
ME-PDG default inputs were used for the crushed granular and CL layers and reflect the values
presented in Table 9.8, these values were also used in the sensitivity analysis
Table 9.8 ME-PDG Inputs for Granular Base and CL Subgrade

Granular Base CL Subgrade
Strength Properties
Poisons ratio 0.27 0.38
Ko 0.37 0.62
Modulus (MPa) 175 110
ICM
Plasticity Index (PI) 1 17
Liquid Limit (LL) 6 18
D60 10.82 0.028
% passing #200 sieve 8.7 70.5
Dry unit weight (kg/m3) 2035 1635
Specific gravity 2.7 2.7
Hydraulic conductivity (mm/hr) 1.55 0.0013
134
Granular Base CL Subgrade
Gravimetric water content (%) 7.4 20
Degree of saturation (%) 61.2 83
9.3.2 Climate
Since no ME-PDG Canadian climatic files exist, some had to be developed. Environment Canada
provided fifteen years of historical climate data (1990-2005) to facilitate the development of a
Southern Ontario (wet-freeze) climatic file. Data for temperature, wind speed, percent sunlight,
precipitation, and relative humidity was gathered to produce the hourly climatic data (HCD) files.
The data provided for temperature, wind speed, percent sunlight, and relative humidity was provided
on an hourly basis. Environment Canada data provided precipitation on a six-hour basis (00:00-
06:00, 06:00-12:00, 12:00-18:00, and 18:00-00:00). However, the ME-PDG requires precipitation
data input on an hourly basis. So in order to modify the data appropriately, eight test climatic files
were prepared to see if allocating the precipitation differently within the sixhour window would
affect the performance results. Precipitation was allocated using:
1. A rectangular distribution,
2. A standard normal distribution N(0,1), and
3. All of the precipitation in one hour. (This was done six times, so each hour within the six-
hour window could be evaluated.)
A graphical representation of the three distributions is presented in Figure 9.9 based on 5 mm of
precipitation over a 6-hour period.
Figure 9.9 Precipitation Distribution Scenarios
135
The results using each of the eigth test climatic files gave exactly the same performance results for
percent slabs cracked, joint faulting, and pavement roughness (IRI).
Since the precipitation distribution did not impact the results the remaining HCD files were
developed for the remaining provinces in Canada using the standard normal N(0,1) distribution. A
sample of the Ontario HCD format is presented below.
[Year][Month][Day][Hour],[Temp.],[Wind Speed],[% Sunlight],[Precip.],[Rel.Humidity]

1990010100,33.98,8.08,40,0,61
1990010101,28.94,3.73,50,0,75
1990010102,30.38,3.73,30,0,79
2005103123,33.08,6.84,0,0,92
The above HCD example summarizes that on January 1, 1990 the temperature at 12 pm was 33.98
o
F, wind was 8.08 MPH, 4 tenths cloud opacity, 0 inches of precipitation and the relative humidity
was 60%. The developed climatic file for Southern Ontario on average would be exposed to 68
freeze-thaw cycles, a freezing index of 505.4oC Day, mean temperature of 8.3 oC and 801.1 mm of
rainfall per year.
9.3.3 Test Section Results and Discussion
The performance results of the ME-PDG simulations at the end of the 50-year analysis period for the
four concrete mixes using the average CTE values are presented in Figures 9.10 through 9.12.
Figure 9.10 Percent Slabs Cracked Performance Results
136
Figure 9.11 Faulting Performance Results
Figure 9.12 Pavement Roughness Performance Results
The following trends were observed from the ME-PDG simulations.

1. All three of the performance indicators (IRI, cracking, and faulting) improved as the
amount of coarse RCA increased. The improvements in performance can be attributed to
the decreasing CTE values as the coarse RCA content increases.
2. The threshold to trigger maintenance and rehabilitation (M&R) is exceeded by the 0%
coarse RCA in percent slabs cracked, faulting, and roughness, and 15% coarse RCA in
faulting and roughness. M&R limits are not reached for 30% and 50% coarse RCA for all
the performance indicators.
137
3. The significant difference in performance for the percentage of slabs cracked between the
0% coarse RCA and the other sections can also be caused by the lower compressive
strength of the test section increasing the rate of load related cracking.
4. Although RCA improved the performance of the pavement in the ME-PDG simulations,
the improvements are not significant in a practical sense. For example an increase in
faulting from 1.75 mm to 3.48 mm is not significant in the field.
To examine the effect of CTE variability on the slab cracking, faulting, and pavement roughness,
twelve additional simulations were completed using the minimum CTE, average CTE or maximum
CTE laboratory values for each coarse RCA amount. The CTE variability results are shown in Table
9.9.
Table 9.9 CTE Variability Results

Slabs Cracked (%)
RCA Amount CTE(Avg.) CTE(Max.) CTE(Min.) Range
0% 38.6 93.6 22.8 70.8
15% 1.2 1.4 0.5 0.9
30% 0.3 0.5 0.2 0.2
50% 1.2 2.3 0.9 1.4
Faulting (mm)
0% 3.48 5.00 2.69 2.31
15% 3.33 3.56 2.21 1.35
30% 2.44 2.82 2.01 0.81
50% 1.75 2.67 1.40 1.27
Roughness - IRI (mm)
0% 3.40 4.60 2.94 1.66
15% 2.82 2.89 2.44 0.45
30% 2.51 2.64 2.37 0.27
50% 2.32 2.63 2.20 0.43
The CTE Variability results show two important trends.
138
1. As the CTE value was increased from the minimum value to the average and then to the
maximum value, IRI, cracking and faulting increased, demonstrating a decrease in
performance. The range in values reinforces the importance of correctly characterizing the
CTE of the concrete.
2. As the amount of RCA increased the range of the results decrease until the 30% RCA
amount and increases again for the three performance indicators.
A summary of the performance results and time of failure for the coarse RCA JPCP test sections
are presented in Table 9.10.
Table 9.10 Summary Results

RCA Content Cracking (% slabs) Faulting (mm) Roughness (m/km)
50-year TOF 50-year TOF 50-year TOF
0% 38.6 38 3.5 44 3.4 39
15% 1.2 NR* 3.3 46 2.8 48
30% 0.3 NR 2.4 NR 2.5 NR
50% 1.2 NR 1.8 NR 2.3 NR
* NR = not reached
139
Chapter 10
Life Cycle Cost Analysis
There is increasing demand on transportation departments to provide longer lasting roads and better
maintenance with decreasing funds and increasing product costs. An effective tool in selecting the
most cost effective design alternatives is life cycle cost analysis (LCCA). LCCA is defined in the
Transportation Equity Act for the 21st Century as a process for evaluating the total economic worth
of a usable project segment by analyzing initial costs and discounted future costs, such as
maintenance, user, reconstruction, rehabilitation, resurfacing costs, over the project segment
[FHWA 2008]. LCCA is not a new concept; it was first proposed for use in transportation projects by
Gillespie in 1847 and became more popular in the 1950s and 60s through the works of Winfrey and
also the American Association of State Highway Officials (AASHO) Red Book [Ozbay 2004].
10.1 LCCA Steps
There are five general steps in conducting a LCCA [Smith 2008].

1. Establish alternative pavement design strategies for the analysis period.
2. Determine performance periods and activity timing.
3. Estimate agency and user costs.
4. Compute life cycle costs.
5. Analyze the results.
10.1.1 Step 1: Establish Alternative Designs
LCCA was used to examine the potential benefit of using three different amounts of coarse RCA
(15%, 30%, and 50%) as a replacement for virgin aggregate using varying pricing scenarios to
simulating the change in cost of virgin aggregate as the supply becomes depleted. Scenario One has
increasing costs as the RCA amount increases, Scenario Two equal costs for all RCA mixes, and
Scenario Three has decreasing costs as the RCA amount increases. The design of the test sections are
based on the CPATT test sections described in Chapter Six and Seven. These include a 250 mm
MTO/OPSS 30 MPa PCC Mix, 100 mm asphalt-stabilized OGDL and a 450 mm granular base.
10.1.2 Step 2: Determine Performance Periods and Activity Timings
A 50-year performance period was selected for the LCCA based on the current practices of the MTO
[Lane 2008, ARA 2007]. Based on the limited empirical performance data for concrete containing
RCA, the standard rigid LCC model used by the MTO was chosen as a starting point. Figure 10.1
presents this model
140
Figure 10.1 Standard Rigid LCCA Model Used
Minor concrete pavement restoration (CPR) at year 18 consists of diamond grinding, joint
resealing, and limited partial- and full-depth patching. Major concrete pavement restoration (CPR) at
year 28 consists of diamond grinding, joint resealing, and limited partial- and full-depth patching. A
90 mm AC overlay at year 38 consists of 40 mm SuperPave (SP) 12.5 mm Friction Course (FC) 2,
and 50 mm SP 19 mm. The activity timings of the MTO rigid LCC model were modified to reflect
concrete material testing results, pavement evaluations, and ME-PDG simulations. The timing of
activities were modeled using a triangular distribution with minimum, most likely, and maximum
values. The timing of the four proposed design alternatives is presented in Table 10.1.
Table 10.1 Proposed Activity Timing Schedules

Activity 0% RCA 15% RCA 30% RCA 50% RCA
Joint Reseal 9, 12, 14 10, 12, 15 11, 12, 16 10, 12, 15
Minor CPR 15, 18, 20 16, 18, 21 17, 18, 22 16, 18, 21
Major CPR 25, 28, 30 26, 28, 31 27, 28, 32 26, 28, 31
90 mm AC Overlay 35, 38, 40 36, 39, 41 37, 40, 42 36, 39, 41
Rout & Seal #1 39, 41, 43 39, 42, 43 39, 43, 45 39, 42, 43
Rout & Seal #2 42, 44, 46 42, 45, 47 43, 46, 48 42, 45, 47
10.1.3 Step 3: Estimate Agency and User Costs
Agency costs include the initial cost of construction and future maintenance and rehabilitation (M&R)
costs minus the remaining value at the end of the design period. User costs include travel time,
vehicle operation costs, accidents, discomfort and user delay. The cost information used in this
LCCA is based upon a study conducted by ARA in 2006 for the Ontario Ministry of Transportation
141
(MTO), Cement Association of Canada (CAC), and Ontario Hot Mix Producers Association
(OHMPA) where LCCA procedures were reviewed and updated to reflect current trends in freeway
pavement designs in Ontario.
In order to estimate the cost in the present, an escalation factor is used to estimate the current dollar
value based on historical costs, using the transportation component of the Consumer Price Index
(CPI) as recommended in the FHWA LCCA Technical Bulletin [Walls III 1998]. The escalation
factor is calculated as 1.037 using Equation 10.1. The transportation component accounts for 19.88%
of the overall CPI.
2008 CPI(Trans)
Escalation Factor = (10.1)
2006 CPI(Trans)
where: CPI(Trans) = Transportation component of the Consumer Price Index
!
The 2008 material and activity costs used in this LCCA were modeled using normal distributions.
! are presented in Table 10.2.
The values
Table 10.2 Material and Activity Costs

Unit Cost St.Dev.
MTO/OPSS 30 MPa PCC Mix $191.00 /m3 $13.44 /m3
Partial Depth PCC Patching $191.32 /m2 $21.32 /m2
Full Depth PCC Patching $174.02 /m2 $32.57 /m2
Diamond Grinding $5.96 /m2 $0.62 /m2
Joint Reseal $5.84 /m $0.72 /m
Superpave 19 mm $80.65 /t $15.74 /t
Superpave 12.5 mm FC2 $109.81 /t $16.76 /t
2
Mill 90 mm HMA $2.16 /m $1.04 /m2
Mill and Patch $15.56 / m2 $1.56 /m2
Rout & Seal $2.75 /m2 $0.21 /m2
Tack Coat $0.32 /m2 $0.09 /m2
User costs were not included in this LCCA, since the MTO and the majority of State DOTs do not
incorporate user costs into their LCCA procedure [Lane 2008, Rangaraju 2008].
142
10.1.4 Step 4: Compute Life Cycle Costs
Probabilistic LCCA was preformed using a Latin Hypercube sampling (LHS) to calculate the Present
Worth (PW) of each design alternative. In order to estimate the PW, 500 simulations with five
iterations per variable were run to calculate an average. LHS is a process where the cumulative
distribution function (cdf) of a specific variable is divided into equal intervals, within each interval a
value is selected randomly, each interval is only sampled once during the simulation. Figure 10.2
shows a simple example of a triangular distribution (3,5,6) that uses five iterations.
Figure 10.2 Triangle Distribution with Five Iterations
The width of intervals for the corresponding value becomes smaller as the most likely value is
approached. Similarly the width of the interval increases as the probability decreases towards the
tails of the distribution. This results in LHS requiring less iteration to achieve convergence than the
Monte Carlo simulation and incorporates low probability values at both ends of the distribution
[Walls III 1998]. PW is calculated using equation 10.2
N " " 1 %% " 1 %

PW = Initial Cost + ($$ M & R Cost$$ ''' ) RV $$ '
nt ' (10.2)
nt '
t=1 # # (1+ i) && # (1+ i) &
where: M & R = maintenance and rehabilitation
RV = remaining value
!
i = discount rate
!
n t = year of expenditure
!
The PW calculation uses a remaining value (RV) at the end of the LCCA period instead of a
! salvage value (SV). A SV represents the cost associated with reclaiming and recycling the materials
! 143
in the road at the end of the serviceable life where as a RV does not account for this [Smith 2008].
The RV is the depreciated value of the structural and functional activity values at the end of the
analysis period. Figures 10.3 and 10.4 illustrate the structural deterioration of the original structure
and functional deterioration of the M&R individually and added together to calculate the RV.
Figure 10.3 Cost Associated with Structural and Functional Deterioration
Figure 10.4 Combined Costs
To model the discount rate, a triangular distribution was used with minimum, most likely, and
maximum values as three percent, five percent, and six percent.
10.1.5 Step 5: LCCA Results Scenario 1 Higher Costs for RCA Concretes
The results of the LHS probabilistic LCCA are presented in Table 10.3 and Figure 10.5.
144
Table 10.3 Results of LCCA Scenario 1
Average $1,088,000 $1,097,000 $1,101,000 $1,117,000
Standard Deviation $23,000 $22,000 $24,000 $23,000
Figure 10.5 Results of LCCA Scenario 1
Several observations were made from the LCCA results.

PW follows the same order as construction cost (0% is the lowest and 50% coarse RCA is
the highest).
The spread between the first and third quartiles (25% and 75%) remains constant for all the
alternatives.
The mean and median values are located very close to the centre 50% percent of the data
showing very little skew to the distribution.
The location of the box within the minimum and maximum values is shifted very slightly
towards the minimum value, indicating a positive skew.
The symmetry of the quartile plots suggests the LCCA PW results follow a normal
distribution.
Figure 10.6 shows the cumulative distribution probabilities of the four design alternatives.
145
Figure 11.6 Cumulative Distribution Probabilities Scenario 1
Using Figure 11.6, the 95% confidence interval (CI) for the LCCA PW mean can be estimated as
follows.
$1,042,000 and $1,135,000 for 0% coarse RCA
Based on the 95% CI values there is no significant difference in the LCCA PW of the four design
alternatives containing varying amounts of coarse RCA.
10.1.6 Step 5: LCCA Results Scenario 2 Equal Costs for RCA and Conventional
Concretes

Average $1,087,000 $1,086,000 $1,083,000 $1,087,000
146

PW follows the order as expected pavement performance (30%, 15%, 50%, and 0% RCA).
alternatives.
Variable skewing between the RCA contents. Slight positive skew for 0% and 30%, a
more positive skew for 15% and a negative skew for the 50%.
distribution with some minor deviations.

147
Using Figure 10.8 the 95% confidence interval (CI) for the LCCA PW mean can be estimated as
follows.
10.1.7 Step 5: LCCA Results Scenario 3 Higher Costs for Conventional Concretes

Average $1,117,000 $1,107,000 $1,093,000 $1,086,000
148
PW follows the same order as construction cost (50% is the lowest and 0% is the highest).
alternatives.
Variable skewing between the RCA contents. Slight positive skew for 0% and 50%, and a
negative skew for 15% and 30%.
distribution with some minor deviations.
Using Figure 10.10, the 95% confidence interval (CI) for the LCCA PW mean can be estimated as
follows.
149
10.2 Sensitivity
10.2.1 Convergence
Figure 10.11 show the convergence results for the PW RCA alternatives.
Figure 10.11 Convergence Results
Very little change in the average LCCA PW is observed in all four design alternatives after 200
simulations. This indicates that the LHS methodology, using 500 simulations with five iterations, is
sufficient to arrive at a meaningful result.
10.2.2 Input Parameters
To estimate the sensitivity of the input parameters a PB screening design, the same as used in Chapter
9, was performed. The eight JPCP factors examined and their values are presented in Table 10.6.
Table 10.6 Value of Factors in LCCA

LCCA Factors - + 0
A Construction Cost 809040 1213560 1011300
B Initial Construction Life 36 40 38
C Discount Rate (%) 3.6 5.6 4.6
D CPR Minor Cost 47559.20 71338.80 59449.00
E CPR Major Cost 59344.16 89016.24 74180.20
F AC Overlay Cost 146761.04 220031.56 183396.30
150
G Quantities
Joint Resealing 280 420 350
Diamond Grinding 6800 10200 8500
Partial Depth Patching 8 12 10
Full Depth Patching 12 18 15
Diamond Grinding 6800 10200 8500
Partial Depth Patching 28 42 35
Full Depth Patching 44 66 55
SP 12.5 mm FC2 708 1062 885
SP 19 mm 832 1248 1040
Tack Coat 6800 10200 8500
Rout & Seal 280 420 350
Rout & Seal 560 840 700
H Remaining Value 29560.21 44340.31 36950.26
The results of the 17 runs are presented in Table 10.7.
Table 10.7 Present Worth Sensitivity Results

Run Variable LCCA PW
A B C D E F G H ($100,000)
1 - - - + + + - + $9.110
2 + - - - - + + + $13.410
3 - + - - + - + + $9.190
4 + + - + - - - + $12.856
5 - - + + - - + + $8.828
6 + - + - + - - + $12.597
7 - + + - - + - + $8.531
8 + + + + + + + + $13.006
9 + + + - - - + - $12.721
10 - + + + + - - - $8.604
151
Run Variable LCCA PW
A B C D E F G H ($100,000)
11 + - + + - + - - $12.710
12 - - + - + + + - $8.930
13 + + - - + + - - $13.011
14 - + - + - + + - $9.449
15 + - - + + - + - $13.482
16 - - - - - - - - $8.775
17 0 0 0 0 0 0 0 0 $10.919
The values from the PB testing ranged from $853,100 to $1,348,200. Using the results from the 17
runs, the effect of the eight variables was plotted against their expected values (calculations are in
Chapter 9) to determine the significant inputs. Figure 10.12 shows the results.
Figure 11.12 Significant Inputs

Analysis of the PB designs show that there are three sensitive input variables in the LCCA model;
construction cost, discount rate, and the quantities.
10.2.3 Discount Rate
To examine the impact the discount rate has on the PW a deterministic LCCA was performed on the
design alternatives using five different rates from two percent to six percent. The results are shown in
Figure 10.13. The discount rate used by the MTO is set by the Ministry of Finance and is currently
set at five percent, prior to 2006 a rate of seven percent was used [Lane 2008, ARA 2007].
152
Figure 10.13 Impact of Discount Rate on Present Worth
As the discount rate increases, the order of PW cost changes. At a discount rate less than three
percent the lowest PW is 30% RCA. When the discount rate is increased between three percent and
four percent, the lowest PW is 15% RCA. Using a discount rate greater than four percent, results in
0% RCA having the lowest PW.
10.3 Discussion
The results of the LCCA show no significant difference between the test sections made with RCA and
the section made without RCA for any of the three scenarios. However, increasing the price of the
virgin aggregate through the three scenarios shows a cost benefit of using the RCA.
The impact on the environment was not considered in the three scenarios. Including the costs
associated with the impact of aggregate pits on plants, animals, noise and water supply, along with the
cost of associated trucking and pollution would further highlight the benefits to using the RCA.
153
Chapter 11
Conclusions and Recommendations
11.1 Conclusions
This research examined the properties of RCA concrete and its performance in JPCP. Compared to a
conventional concrete mixture, substituting RCA for virgin coarse aggregate resulted in similar or
improved performance. Indicating that RCA is a viable aggregate source for concrete pavements.
Material testing of the RCA used in the preliminary mixes and placed at the CPATT test track
showed that the source would have been classed as unacceptable based on the MTOs virgin material
specification OPSS 1002. Compared to the corresponding nineteen millimeter virgin aggregate, the
RCA had an increase in absorption of 300% and decreases in bulk relative density of thirteen percent
and abrasion resistance of 34% measured using Micro-Deval testing.
The compressive strength of the twelve preliminary design mixes at 28-days all exceeded the 30
MPa design strength. For the three coarse RCA amounts (15%, 30%, and 50%) compressive strength
increases were related to the increases in cement content (315 kg/m3, 330 kg/m3, and 345 kg/m3),
however, there was no statistical difference between the mixes at the same cement content.
The compressive strength of the concrete increased as the amount of coarse RCA increased to a
maximum at 30% at 28-days. All three of the coarse RCA mixes demonstrated increases in strength
compared to the conventional concrete. The flexural strength results followed the same trends as the
compressive strength. The compressive and flexural strength values are as follows:
0% coarse RCA 27.6 MPa (Compressive) and 6.0 MPa (Flexural),
15% coarse RCA 40.9 MPa, and 7.2 MPa,
30% coarse RCA 42.1 MPa, and 7.3 MPa, and
50% coarse RCA 35.0 MPa, and 6.3 MPa.
Freeze-thaw (F/T) testing showed no difference between the coarse RCA mixes and the
conventional concrete in respect to change in relative dynamic modulus (RDM) and percent mass
loss. At the end of the 300 cycles, the RDM was approximately 90% and the mass loss was less than
1%. Both the RDM and mass loss decreased as the coarse RCA amount increased up to 30%. The
RDM and mass loss values are as follows:
0% coarse, RCA 91.3% (RDM), and 0.25% (mass loss),
15% coarse RCA 90.8%, and 0.2%,
30% coarse RCA 91.9%, and 0.16%, and
50% coarse RCA 91.1%, and 0.11%.
154
The coefficient of thermal expansion (CTE) decreased as the coarse RCA amount increased and
was in the typical range of a conventional concrete made with limestone aggregate. The CTE values
are as follows:
0% RCA 7.28x10-6 /oC,
15% RCA 6.603x10-6 /oC,
30% RCA 5.265x10-6 /oC, and
50% RCA 4.103x10-6 /oC.
The simplified CTE testing method provided two key benefits in modeling the observed field
conditions over the AASHTO TP 60 method. These included:
The testing temperature range, -20oC to 20oC, and
The specimen condition, not being immersed in water.
After two years in-service and approximately 3,000,000 ESALs, all test sections are in excellent
condition with pavement condition index (PCI) values greater than 85. The sections containing RCA
are showing no significant difference in performance compared to the conventional JPCP section.
The PCI values at the June 2009 evaluation:
0% coarse RCA 91,
15% coarse RCA 95,
30% coarse RCA 94, and
50% coarse RCA 95.
The test sections are all showing the same distresses; abrasion, joint cracking and spalling, and
raveling.
Sensor data from the strain gauges, and vertical and inter-panel extensometers are providing
consistent results between the test sections. 80% of the sensors are still functioning and providing
useful data after two years.
Strain findings:
1. The change in strain is greater for the sensors located in the upper portion of the slab
compared to the bottom.
2. The change in strain is greater in the transverse direction compared to longitudinal.
3. The change in strain is constant for transverse directions located mid slab and at the joint.
Vertical and Inter-Panel extensometer findings:
1. The change in vertical displacement is three times greater at the pavement edge compared
to the pavement centreline.
2. The change in horizontal displacement is equal at the pavement edge compared to the
pavement centreline.
155
Long-term performance modeling of percent cracked slabs, joint faulting, and pavement roughness
using the ME-PDG showed improved performance as the coarse RCA content increased for all
measures. At the end of the 50-year analysis period, only five simulations exceeded the specified
trigger levels for maintenance and rehabilitation to occur. These simulations included: 0% coarse
RCA for percent slabs cracked, joint faulting, and roughness; and 15% coarse RCA for joint faulting
a, and roughness. Multivariable sensitivity analysis showed that the performance results were
sensitive to CTE, unit weight, joint spacing, edge support, surface absorption, and dowel bar
diameter. Additional ME-PDG simulations using the maximum and minimum CTE values showed
improved performance as the CTE value decreased for all coarse RCA amounts.
The three life cycle cost analysis (LCCA) showed cost savings could be achieved by using RCA as
virgin aggregate resources diminish. For each of the scenarios no significant difference between the
coarse RCA amounts was observed on a per kilometer basis, however, over the length of an entire
roadway the savings become significant.
Scenario 1 Abundant supple of virgin aggregate, higher costs for RCA concretes.
0% RCA $1,088,000/km,
15% RCA $1,097,000/km,
30% RCA $1,101,000/km, and
50% RCA $1,117,000/km.
Scenario 2 Diminishing supply of virgin aggregate, equivalent cost for RCA and conventional
concretes.
0% RCA $1,087,000/km,
15% RCA $1,086,000/km,
30% RCA $1,083,000/km, and
50% RCA $1,087,000/km.
Scenario 3 Shortage of virgin aggregate, lowers costs for RCA concretes.
0% RCA $1,117,000/km,
15% RCA $1,107,000/km,
30% RCA $1,093,000/km, and
50% RCA $1,086,000/km.
Multivariable sensitivity analysis showed that the LCCA results were sensitive to construction
costs, discount rate, and maintenance and rehabilitation quantities. Additional LCCA, using the
variable discount rates between two percent and six percent, changed the order of the present worth
values.
156
11.2 Recommendations
For RCA to become widely used material, consistent and predictable results need to be obtained when
using as a substitute for virgin aggregate in concrete. To achieve this further study is required in the
areas of aggregate properties, mixture design and proportioning, performance, testing, and modeling.
Aggregate Properties
Perform petrographic analysis on the RCA samples to better understand their composition,
quality, and how much deleterious material that can be included without affecting the
performance of the concrete.
Mixture Design and Proportioning
Comparing concrete mixes with different sources of RCA including sources of RCA that
are clean, contaminated, and cured differently.
Compare concrete mixes with a variety of coarse RCA content to find the optimal amount
that can be added without sacrificing performance.
Develop additional designs that incorporate fine RCA and concrete wash-water to achieve
a zero waste concrete.
Performance
Continued monitoring of pavement structure (PCI surveys and sensor data) to understand
the long-term performance.
Testing
Retest RCA cylinders using AASHTO TP-60 method to compare and validate results.
Based on the results it may be possible to conclude that the simplified method is acceptable
for calculating CTE for use in the ME-PDG.
Investigate the impact of core condition (dry, saturated surface dry, wet) has on the CTE
using simplified method.
Modeling
Estimate pavement performance of concrete containing RCA in additional climatic zones
(wet no-freeze, dry-freeze, and dry no-freeze) and severities using the ME-PDG to
determine its suitability in those locations.
Update PCI, sensor, and LCCA models as new data becomes available.
157
Bibliography
AASHTO. (2004). Standard Test Method for Coefficient of Thermal Expansion of Hydraulic
Concrete, AASHTO TP-60, American Association of State Highway and Transportation
Officials, Washington, D.C.
Abbas, A., Fathifazl, G., Isgor, O.B., Razapur, A.G., Fournier, B., and Foo, S. (2008). Durability of
Recycled Aggregate Concrete Designed with Equivalent Mortar Volume (EVM) Method, 2nd
Canadian Conference on Effective Design of Structures, paper #58, McMaster University,
Hamilton, ON.
Abou-Zeid, M.N., Shenouda, M.N., McCabe, S.L., and El-Tawil, F.A. (2005). Reincarnation of
Concrete, Concrete International, V. 27, No.2, February 2005, pp. 53-59.
Achtemichuk, S, Hubbard, J, Sluce, R., and Shehata, M. (2008). Development of Controlled Low-
Strength Materials Using Reclaimed Concrete Aggregates, 2nd Canadian Conference on
Effective Design of Structures, paper #58, McMaster University, Hamilton, ON, 2008.
American Concrete Institute. (2005). Building Code Requirements for Structural Concrete and
Commentary, ACI 318-05, Farmington Hills, MI, 2005, pp. 1-430.
American Concrete Institute. (1992). State-of-the-Art on High-Strength Concrete, ACI 363R-92,

Farmington Hills, MI, 1992, pp. 1-55.
American Concrete Pavement Association. (2009). Recycling Concrete Pavement, Concrete

Paving Technology, EB-043P.
American Concrete Pavement Association. (1993). Recycling Concrete Pavement, Concrete

Paving Technology, TB-014P.
Ajdukiewicz, A., and Kliszczewica, A. (2002). Influence of Recycled Aggregates on Mechanical

Properties of HS/HPS, Cement and Concrete Composites, V. 24, No. 2, 2002, pp. 269-279.
Aggregate Producers Association of Ontario. (2004). Importance of Aggregate, The About

Aggregate Series, Aggregate Producers Association of Ontario, Mississauga, Ontario, pp. 1-4.
Applied Research Associates, Inc. (2007). Life-Cycle Cost 2006 Update, Toronto, ON.
158
Applied Research Associates, Inc., ERES Consultants Division. (2004). Guide for Mechanistic-
Empirical Design of New and Rehabilitated Pavement Structures, NCHRP Project 1-37A,
National Cooperative Highway Research Program, Washington, D.C.
ASTM. (2004). Annual Book of ASTM Standards, Volume 04.02, American Society for Testing
and Materials, West Conshohocken, PA.
Bairagi, N. K., Vidyadhara, H. S., and Ravande, K. (1990). Mix Design Procedure for Recycled
Aggregate Concrete, Construction and Building Materials, V. 4, No. 4, December 1990, pp. 188-
193.
Kosmatka, S.H., Kerhoff, B., MacLeod, N.F., and McGrath, R.J. (2002). Design and Control of
Concrete Mixtures, EB101, 7th Canadian Edition, Cement Association of Canada, Ottawa, ON.
Chen, H.J., Yen, T., and Chen, K.H. (2003). Use of Building Rubbles as Recycled Aggregate,
Cement and Concrete Research, V.33, No.1, pp. 125-132.
Chini, A.R., Kuo, S.S., Armaghani, J.M., and Duxbury, J.P. (2001). Test of Recycled Concrete
Aggregate in Accelerated Test Track, Journal of Transportation Engineering, V.127, No.6,
November/December 2001, pp. 486-492.
Chong, G.J., and Wrong, G.A. (2001). Manual for Condition Rating of Rigid Pavements
Concrete Surface and Composite Distress Manifestations, SP-026, Ministry of Transportation,
Ontario, Downsview, ON.
Craig, R.F. (1997). Soil Mechanics, Sixth Edition, E&FN Spon, London, UK.
CSA International. (2000). Concrete Materials and Methods of Concrete Construction, CSA
A23.1-00, Canadian Standards Association, Mississauga, ON.
Cuttell, G.D., Snyder, M.B., Vandenbossche, J.M., and Wade, M.J. (2008). Performance of Rigid
Pavements Containing Recycled Concrete Aggregate, Transportation Research Record 1574,
National Academy of Science, Washington, DC, 1997.
Darter, M., Khazanovich, L., Yu, T., and Mallela, J. (2005). Reliability Analysis of Cracking and
Faulting Predection Included in the New Mechanistic-Empirical Pavement Design Procedure, 05-
1917, Transportation Research Board, 84th Annual Meeting, Washington, DC, January 9-13.
159
Dore, G., Tighe, S., and Hein, D. (2005). Roadmap for the Calibration and Implementation of the
Mechanistic-Empirical Design of New and Rehabilitated Pavement Structures, Transportation
Association of Canada, Ottawa, ON.
Dosho, Y. (2005). Application of Recycled Aggregate Concrete for Structural Concrete: A

Recycling System for Concrete Waste, International Symposium on Sustainable Development of
Cement, Concrete and Concrete Structures, Toronto, Ontario, October 5-7, pp. 459-478.
Farhey, D.N. (2007). Instrumentation Specifics: Choices and Trade-Offs, Transportation Research
Board, 86th Annual Meeting, Washington, DC, January 21-25.
Fathifazl, G., Razapur, A.G., Isgor, O.B., Abbas, A., Fournier, B., and Foo, S. (2008). Fresh and
Hardened Properties of Recycled Aggregate Concrete Proportioned by the Equivaletn Mortar
Volume (EMV) Method, 2nd Canadian Conference on Effective Design of Structures, paper #95,
McMaster University, Hamilton, ON.
FHWA. (2008). TEA-21 Moving Americans into the 21st Century, Federal Highways
Administration, www.fhwa.dot.gov/Tea21/index.htm accessed July 2008.
FHWA. (2007). FHWA Recycled Materials Policy, Federal Highways Administration,
www.fhwa.dot.gov/legsregs/directives/policy/recmatpolicy.htm, accessed July 2007.
FHWA. (2005). DataPave Online, U.S. Department of Transportation, Federal Highways

Administration, http://www.datapave.com, accessed May 14, 2005.
FHWA. (2004). Transportation Applications Of Recycled Concrete Aggregate: FHWA State of the
Practice National Review September 2004, U.S. Department of Transportation, Federal
Highways Administration, Washington, DC.
FHWA. (1990). Concrete Pavement Joints, Technical Advisory T 5040.30, U.S. Department of
Transportation, Federal Highways Administration, Washington, DC.
Khaled A. Galal, K.G., and Chehab, G.R. (2005). Considerations for Implementing the 2002 M-E
Design Procedure Using a HMA Rehabilitated Pavement Section in Indiana, 05-1888,
Transportation Research Board, 84th Annual Meeting, Washington, DC, January 9-13.
Gilbert, N. (2005). Personal email correspondence, Ontario Ministry of Transportation, Downsview,

ON, January 5, 2005.
160
Gokce, A., Nagataki, S., Saeki, T., and Hisada, M. (2004). Freezing and Thawing Resistance of
Air-Entrained Concrete Incorporating Recycled Coarse Aggregate: The Role of Air Content in
Demolished Concrete, Cement and Concrete Research, V. 34, No. 5, pp. 799-806.
Gomez-Soberon, J.M.V. (2002). Porosity of Recycled Concrete with Substitution of Recycled

Concrete Aggregate An Experimental Study, Cement and Concrete Research, V.32, No.8, pp.
1301-1311.
GTAA. (2007). Reducing, Reusing and Recycling Terminal 2, Toronto Pearson Today: Terminal
2, Terminal 2 Commemorative Issue, Greater Toronto Airports Authority, Toronto, ON.
Hall, K.D., and Beam, S. (2005). Estimation of Sensitivity of Design Variables for Rigid Pavement
Analysis Using the Mechanistic-Empirical Design Guide, 05-0948, Transportation Research
Board, 84th Annual Meeting, Washington, DC, January 9-13.
Hansen, T.C. (1990). Recycled Concrete Aggregate and Fly Ash Produced Concrete Without
Portland Cement, Cement and Concrete Research, V. 20, No.3, pp. 355-356.
Hansen, T.C., and Angelo, J.W. (1986). Crushed Concrete Fines Recycled for Soil Modification
Purposes, ACI Materials Journal, V. 83, No. 6, November-December 1986, pp. 983-987.
Hansen, T.C., and Boegh, E. (1985). Elasticity and Drying Shrinkage of Recycled-Aggregate
Concrete, ACI Materials Journal, V. 82, No. 5, September-October 1985, pp. 648-652.
Hansen, T.C., and Hedegard, S.E. (1984). Properties of Recycled Aggregate Concretes as Affected
by Admixtures in Original Concretes, ACI Journal, January-February 1984, pp. 21-26.
Torben C. Hansen, T.C., and Narud, H. (1983). Recycled Concrete and Silica Fume Make Calcium
Silicate Bricks, Cement and Concrete Research, V. 13, No.5, pp. 626-630.
Harrington, J. (2004). States Achieve Recycling Success, Roads and Bridges, V.42, No.7.
Kamel, A.M.E.A., and Abou-Zeid, M.N. (2008). Key Performance and Economic Considerations
for Using Recycled Concrete Aggregate, 87th Annual Meeting, Transportation Research Board,
National Academy of Science, 08-1073, Washington, DC.
Katz A. (2004). Treatments for the Improvement of Recycled Aggregate, Journal of Materials in
Civil Engineering, V. 16 No. 6, November/December 2004 pp. 597-603.
161
Katz, A. (2003). Properties of Concrete Made with Recycled Aggregate from Partially Hydrated
Old Concrete, Cement and Concrete Research, V. 343 No. 5, pp. 703-711.
Kerkhoff, B., and Siebel, E. (2001). Properties of Concrete with Recycled Aggregates, Beton
2/2001, Verlag Bau + Technik, pp. 105-108.
Kuroda, Y., and Hashida, H. (2005). A Closed-loop Concrete System on a Construction Site,
International Symposium on Sustainable Development of Cement, Concrete and Concrete
Structures, Toronto, Ontario, October 5-7, pp. 371-388.
Lane, B. (2008). Life Cycle Costing and Alternative Bid Contracts in Ontario, Canada,
Transportation Research Board, 87th Annual Meeting, National Academy of Science, Washington,
DC, January 13-17.
Li, X., and Gress, D.L. (2006). Mitigating Alkali-Silica Reaction in Concrete Containing Recycled
Concrete Aggregate, Transportation Research Record 1979, National Academy of Science,
Washington, DC.
Li, S., Nantung, T., and Jiang, Y. (2005). Assessing Issues, Technologies, and Data Needs to Meet
Traffic Input Requirements by M-E Pavement Design Guides: Implementation Initiatives, 05-
1431, Transportation Research Board, 84th Annual Meeting, Washington, DC, January 9-13.
Lin, Y.H., Tyan, Y.Y., Chang, T.P., and Chang, C.Y. (2004). An Assessment of Optimal Mixture
for Concrete Made With Recycled Concrete Aggregates, Cement and Concrete Research, V. 34,
No. 8, pp. 1373-1380.
Liu, J., Zollinger, D.G., Tayabji, S.D., and Smith, K.D. (2005). Application of Reliability Concept
in Concrete Pavement Rehabilitation Decision Making, 05-2202, Transportation Research Board,
84th Annual Meeting, Washington, DC, January 9-13.
Masood, A., Ahmad, T., Arif, M., and Mahdi, F. (2002). Waste Management Strategies for
Concrete Environmental Engineering and Policy, V.3, pp. 15-18.
MacNaughton, I. (2004). Ontarios Aggregates Are a Diminishing Resource, Ontario Hot Mix
Producers Association Fall Seminar Presentations, December 8.
Mehta, P.K., and Monteiro, P.J.M. (2005). Concrete: Microstructure, Properties, and Materials, 3rd
Edition, McGraw-Hill Companies Inc., New York, NY.
162
Mehta, P.K. (2001). Reducing the Environmental Impact of Concrete, Concrete International, V.
23, No.10, October 2001, pp. 61-66.
Mehta, P.K. (1999). Concrete Technology for Sustainable Development, Concrete International,
V. 21, No.11, November 1999, pp. 47-53.
Mesbah, H.A., and Buyle-Bodin, F. (1999). Efficiency of Polypropylene and Metallic Fibres on
Control of Shrinkage and Cracking of Recycled Aggregate Mortars, Construction and Building
Materials, V. 13, No. 8, pp. 439-447.
Meyer, C. (2008). The Greening of the Concrete Industry, 2nd Canadian Conference on Effective
Design of Structures, paper #97, McMaster University, Hamilton, ON, 2008.
Miller, G. (2005). Running Out of Gravel and Rock, Toronto Star, January 6, p. A22.
Douglas C. Montgomery, D.C. (2001). Design and Analysis of Experiments, 5th Edition, John
Wiley & Sons Inc., New York, NY.
MTO. (2004). Laboratory Testing Manual, Ontario Ministry of Transportation, Downsview, ON.
Nagataki, S., Gokce, A., Saeki, T., and Hisada, M. (2004). Assessment of Recycling Process
Induced Damage Sensitivity of Recycled Concrete Aggregates, Cement and Concrete Research,
V. 34, No. 6, pp.965-971.
Naik, T.R., and Moriconi, G. (2005). Environmental-Friendly Durable Concrete Made with
Recycled Materials for Sustainable Concrete Construction, International Symposium on
Sustainable Development of Cement, Concrete and Concrete Structures, Toronto, Ontario,
October 5-7, pp. 277-298.
Naik, T.R. (2002). Greener Concrete Using Recycled Materials, Concrete International, V. 24,
No.7, July 2001, pp. 45-49.
Nantung, T., Chehab, G., Newbolds, S., Galal, K., Li, S. and Kim, D.H. (2005). Implementation
Initiatives of the Mechanistic-Empirical Pavement Design Guides in Indiana, 05-1407,
Transportation Research Board, 84th Annual Meeting, Washington, DC, January 9-13.
Neville, A.M. (1995). Properties of Concrete, Fourth Edition, Prentice Hall, Essex, England.
163
Noguchi, T. (2005). An Outline of Japanese Industrial Standards (JIS) as Related to Sustainability
Issues, International Symposium on Sustainable Development of Cement, Concrete and Concrete
Structures, Toronto, Ontario, October 5-7, 2005, pp. 407-422.
Karthik Obla, K., Kim, H., and Lobo, C. (2007). Crushed Returned Concrete as Aggregates for
New Concrete Final Report to the RMC Research & Education Foundation, Project 05-13,
National Ready Mixed Concrete Association.
Oikonomou, N.D. (2005). Recycled Concrete Aggregates, Cement and Concrete Composites, V.
25, No. 2, pp. 315-318.
Olorunsogo, F.T., and Padayachee, N. (2002). Performance of Recycled Aggregate Concrete

Monitored by Durability Indexes, Cement and Concrete Research, V. 32, No. 2, 2002, pp. 179-
185.
Ontario Provincial Standard Specification. (2005). Material Specification for Aggregate

Concrete, OPSS 1002 (Draft 2), Ontario Ministry of Transportation, St. Catherines, ON.
Otsuki, N., Miyazato, S.I., and Yodsudjai, W. (2003). Inluence of Recycled Aggregate in
Interfacial Transition Zone, Strength, Chloride Penetration and Carbonation of Concrete, Journal
of Materials in Civil Engineering, V.15, No.5, September/October 2003, pp. 443-451.
Ozbay, K., Jawad, D., Parker, N.A., and Hussain, S. (2004). Life Cycle Cost Analysis: State-
of0the-Practice vs State-of-the -Art, 04-3115, Transportation Research Board, 83rd Annual
Meeting, Washington, DC, January 11-15.
Poon, C.S., Shui, Z.H. and Lam, L. (2004). Effect of Microstructure of ITZ on Compressive
Strength of Concrete Prepared with Recycled Aggregate, Construction and Building Materials,
V. 18, No. 6, pp. 461-468.
Poon, C.S., Shui, Z.H., Lam, L., Fok, H., and Kou, S.C. (2003). Influence of Moisture States of
Natural and Recycled Aggregates on the Slump and Compressive Strength of Concrete, Cement
and Concrete Research, V.34, No.1, 2004, pp. 31-36.
Poon, C.S., Kou, S.C., and Lam, L. (2002). Use of Recycled Aggregates in Molded Concrete Bricks
and Blocks, Construction and Building Materials, V. 16, No. 5, pp. 281-289.
Rangaraju, P.R., Amirkhanian, S., and Guven, S. (2008). Life Cycle Cost Analysis for Pavement
Type Selection, SCDOT Final Report, FHWA-SC-08-01, Clemenson, SC, 2008.
164
Rashwan, M.S., and AbouRizk, S. (1997). The Properties of Recycled Concrete, Concrete
International, V. 18, No.7, July 1997, pp. 56-60.
Ryu, J.S. (2002). Improvement on Strength and Impermeability of Recycled Concrete Made from
Crushed Concrete Coarse Aggregate, Journal of Materials Science Letters, V.21, pp. 1565-1567.
Saeed, A. (2006). Evaluation, Design and Construction Techniques for Airfield Concrete pavement
Used as Recycled Material for Base, Innovative pavement Research Foundation, IPRF-01-G-
002-03-5, Skokie, IL.
Sagoe-Crentsil, K.K., Brown, T., and Taylor, A.H. (2001). Performance of Concrete Made with
Commercially Produced Coarse Recycle Concrete Aggregate, Cement and Concrete Research, V.
31, No. 5, pp. 707-712.
Salem, R.M., Burdette, E.G., and Jackson, N.M. (2003). Resistance to Freezing and Thawing of
Recycled Aggregate Concrete, ACI Materials Journal, V. 100, No. 3, May-June 2003, pp. 216-
221.
Salem, R.M., and Burdette, E.G. (1998). Role of Chemical and Mineral Admixtures on Physical
Properties and Frost-Resistance of Recycled Aggregate Concrete, ACI Materials Journal, V. 95,
No. 5, September-October 1998, pp. 558-563.
Seow, K. (2005). Old Terminal 1 Decommissioning and Demolition, Annual Conference of the
Transportation Association of Canada, Calgary, AB, 2005.
Shayan A., and Xu, A. (2003). Performance and Properties of Structural Concrete Made with
Recycled Concrete Aggregate, ACI Materials Journal, V. 100, No. 5, September-October 2003,
pp. 371-380.
Smith, M. (2008). Life-Cycle Cost Analysis in Pavement Type Selection, Wisconsin Asphalt
Pavement Association Conference, Atlanta, GA.
Smith, J.T., and Tighe, S.L. (2004). Assessment of Overlay Roughness in the LTPP: A Canadian
Case Study, 04-2502, Transportation Research Board Conference Proceedings, Washington, D.C.
Soilmoisture. (1998). Concrete and Trase TDR, Product Information, Soilmoisture Equipment
Corp., Santa Barbara, CA.
St Marys Cement. (2007). General Use Portland Cement, Product Information, St Marys Cement
Group, Toronto, ON.
165
TAC. (1997). Pavement Design and Management Guide, Transportation Association of Canada,
Ottawa, ON.
Tanesi, J., Kutay, M.E., Abbas, A., and Meininger, R. (2007). Effect of CTE Variability on
Concrete Pavement Performance as Predicted Using the Mechanistic-Empirical Pavement Design
Guide, Transportation Research Board, 86th Annual Meeting, Paper No. 07-2488, Washington,
D.C.
Tavakoli, M. and Soroushian, P. (1996a). Strengths of Recycled Concrete Made Using Field-
Demolished Concrete as Aggregate, ACI Materials Journal, V. 93, No. 2, March-April 1996, pp.
182-190.
Tavakoli, M. and Soroushian, P. (1996b). Drying Shrinkage Behavior of Recycled Aggregate

Concrete, Concrete International, V. 18, No.11, November 1996, pp. 58-61.
Topcu, I.B., and Sengel, S. (2003). Properties of Concretes Produced with Waste Concrete
Aggregate, Cement and Concrete Research, V.34, No.8, pp.1307-1312.
Topcu, I.B. (1997). Physical and Mechanical Properties of Concrete Produced with Waste
Concrete, Cement and Concrete Research, V. 27, No.12, pp. 1817-1823.
Topcu, I.B., and Guncan, N.F. (1995). Using Waste Concrete as Aggregate, Cement and Concrete
Research, V. 25, No.7, 1995, pp. 1385-1390.
Transtec. (2005). HYPERPAV II, Software Version 3.00.0061, The Transtec Group, Austin, TX.
Turley, W. (2003). On the road: acceptance of recycled aggregates by state DOTs is slow but
making progress Roadbuilding Trends, C&D Recycler, September-October, 2003.
USGBC. (2005). LEED for New Construction, Version 2.2, U.S. Green Building Council,
Washington, DC.
Walls III, J., and Smith, M.R., (1998). Life-Cycle Cost Analysis in Pavement Design Interim
Technical Bulletin, FHWA-SA-98-097, Washington, DC.
West, J. (2005). Effect of Temperature on Structural Performance of Stainless Steel Reinforcement,

Final Report for the Ontario Ministry of Transportation, University of Waterloo, ON.
166
Yang, S., Kim, N., Kim, J., and Park, J. (2003). Experimental Measurement of Concrete Thermal
Expansion, Journal of the Eastern Asia Society for Transportation Studies, V.5, October 2003,
pp. 1035-1048.
Yu, H.T., and Khazanovich, L. (2005). Use of Magnetic Tomography to Evaluate Dowel
Placement, FHWA-IF-06-006, Federal Highways Administration, Office of Pavement
Technology, Washington, D.C.
Zaharieva, R., Buyle-Bodin, F., and Wirquin, E. (2004). Frost Resistance of Recycled Aggregate
Concrete, Cement and Concrete Research, V. 34, No.10, pp. 1927-1932.
Zaharieva, R., Buyle-Bodin, F., Skoczylas, F., and Wiequin, E. (2003). Assessment of the Surface
Properties of Recycled Aggregate Concrete, Cement and Concrete Composites, V. 25, No. 2, pp.
223-232.
167
Appendix A
Design Tables
CBR = 2.5, Asphalt Stabilized OGDL

Traffic ADTT Concrete Flexural Strength (MR)
(2-way) 3.5 MPa 5.0 MPa 6.5 MPa 8.0 MPa
Residential 2 127 101.6 101.6 101.6
10 139.7 114.3 101.6 101.6
25 139.7 114.3 101.6 101.6
Collector 100 165.1 127 114.3 101.6
200 165.1 139.7 114.3 101.6
400 165.1 139.7 114.3 114.3
Minor Arterial 500 190.5 152.4 139.7 127
1000 190.5 165.1 139.7 127
1500 203.2 165.1 139.7 139.7
Major Arterial 2000 215.9 177.8 152.4 139.7
3000 215.9 177.8 152.4 152.4
5000 215.9 177.8 165.1 152.4
CBR = 2.5, Cement Treated OGDL

Residential 2 127 101.6 101.6 101.6
10 139.7 114.3 101.6 101.6
25 139.7 114.3 101.6 101.6
Collector 100 152.4 127 114.3 101.6
200 165.1 127 114.3 101.6
400 165.1 139.7 114.3 114.3
Minor Arterial 500 190.5 152.4 127 127
1000 190.5 152.4 139.7 127
1500 190.5 165.1 139.7 139.7
Major Arterial 2000 215.9 177.8 152.4 139.7
3000 215.5 177.8 152.4 152.4
5000 215.9 177.8 165.1 152.4
168
CBR = 2.5, Non-treated OGDL
Residential 2 127 101.6 101.6 101.6
10 139.7 114.3 101.6 101.6
25 139.7 114.3 101.6 101.6
Collector 100 165.1 139.7 114.3 101.6
200 165.1 139.7 114.3 101.6
400 177.8 139.7 127 114.3
Minor Arterial 500 190.5 165.1 139.7 127
1000 203.2 165.1 139.7 127
1500 203.2 165.1 139.7 139.7
Major Arterial 2000 215.9 177.8 152.4 139.7
3000 215.9 177.8 152.4 152.4
5000 228.6 177.8 165.1 152.4
CBR = 5, Asphalt Stabilized OGDL

Residential 2 127 101.6 101.6 101.6

10 127 101.6 101.6 101.6
25 139.7 114.3 101.6 101.6
Collector 100 152.4 127 114.3 101.6
200 165.1 127 114.3 101.6
400 165.1 127 114.3 114.3
Minor Arterial 500 190.5 152.4 127 114.3
1000 190.5 152.4 139.7 127
1500 190.5 152.4 139.7 127
Major Arterial 2000 203.2 165.1 152.4 139.7
3000 215.9 177.8 152.4 139.7
5000 215.9 177.8 152.4 152.4
169
CBR = 7.5, Asphalt Stabilized OGDL
Residential 2 114.3 101.6 101.6 101.6
10 127 101.6 101.6 101.6
25 139.7 114.3 101.6 101.6
Collector 100 152.4 127 101.6 101.6
200 152.4 127 114.3 101.6
400 165.1 127 114.3 101.6
Minor Arterial 500 177.8 152.4 127 114.3
1000 190.5 152.4 127 127
1500 190.5 152.4 139.7 127
Major Arterial 2000 203.2 165.1 139.7 139.7
3000 203.2 165.1 152.4 139.7
5000 203.2 165.1 152.4 152.4

Residential 2 114.3 101.6 101.6 101.6
10 127 101.6 101.6 101.6
25 127 101.6 101.6 101.6
Collector 100 152.4 127 101.6 101.6
200 152.4 127 101.6 101.6
400 152.4 127 114.3 101.6
Minor Arterial 500 177.8 139.7 127 114.3
1000 177.8 152.4 127 127
1500 190.5 152.4 139.7 127
Major Arterial 2000 203.2 165.1 139.7 139.7
3000 203.2 165.1 152.4 139.7
5000 203.2 165.1 152.4 152.4
170
Residential 2 114.3 101.6 101.6 101.6
10 127 101.6 101.6 101.6
25 127 101.6 101.6 101.6
Collector 100 152.4 114.3 101.6 101.6
200 152.4 127 101.6 101.6
400 152.4 127 114.3 101.6
Minor Arterial 500 177.8 139.7 127 114.3
1000 177.8 152.4 127 127
1500 177.8 152.4 127 127
Major Arterial 2000 203.2 165.1 139.7 139.7
3000 203.2 165.1 139.7 139.7
5000 203.2 165.1 152.4 139.7

Residential 2 114.3 101.6 101.6 101.6
10 127 101.6 101.6 101.6
25 127 101.6 101.6 101.6
Collector 100 139.7 114.3 101.6 101.6
200 152.4 114.3 101.6 101.6
400 152.4 127 101.6 101.6
Minor Arterial 500 177.8 139.7 127 114.3
1000 177.8 139.7 127 114.3
1500 177.8 152.4 127 127
Major Arterial 2000 190.5 152.4 139.7 127
3000 203.2 165.1 139.7 139.7
5000 203.2 165.1 152.4 139.7
171
Residential 2 114.3 101.6 101.6 101.6
10 114.3 101.6 101.6 101.6
25 127 101.6 101.6 101.6
Collector 100 139.7 114.3 101.6 101.6
200 139.7 114.3 101.6 101.6
400 152.4 127 101.6 101.6
Minor Arterial 500 165.1 139.7 114.3 101.6
1000 177.8 139.7 127 114.3
1500 177.8 139.7 127 114.3
Major Arterial 2000 190.5 152.4 139.7 127
3000 190.5 152.4 139.7 127
5000 190.5 165.1 139.7 139.7
172
Appendix B
Preconstruction Meeting Summaries
Meeting #1
March 8, 2007 - UW/CAC/Dufferin Conference Call
Materials: Since stockpiles of material to be recycled often contain both asphalt and concrete, only
pure waste concrete should be set aside for crushing. The minimum amount of aggregate required is
150 tonne, however 200 tonne would be preferred. Potential companies who may be a source of RCA
are Capital paving, Cox Construction, Murray Group, and Steed and Evans.
Design: A 200 mm JPCP was determined to be too thin for the automatic dowel-bar inserter. The
MTO recommends using a 220-250 mm surface since the extra cost is not significant and it will
greatly improve performance. Non-treated open graded drainage layer (OGDL) was recommended
for use as a base. The design would need to be rechecked with the MTO and others.
Equipment: The pavement slab width will depend on the existing setup of the paver. Since it is
difficult to change the setup of the paver for each texture and changes may produce a non-uniform
surface, it is best to use the paver at the existing set up. An adaptor kit will be required for the paver
to produce the desired tine. The adaptor kit can potentially be borrowed from CoCo Paving.
Schedule: The Region of Waterloo preferred that construction take place in May, however, Dufferin
felt that in order to give time for mix development mid- to late-June would be a more realistic time
frame. James from UW will work with Jai from Dufferin to develop the mix designs. The
construction of the test sections can be completed within a week.
Additional Work: The construction of the test sections is a potential opportunity to look at the effect
of dowel bar misalignment. Follow up will be done with the MTO on this.
173
Meeting #2
April 17, 2007 UW/CAC/Dufferin (at Dufferin Construction)
The premise of placing RCA test sections is to showcase new technology and evaluate the
performance.
Materials
Steed and Evans will crush the RCA and donate the material to the project. Another location will
need to be used for screening. The RCA used in the project will originate from used sidewalks and
curbs and will be 4.75 mm to 19mm in size. The RCA will be combined with virgin aggregate of
4.75mm to 40mm.
Design
Four 50 m test sections containing varying amounts of RCA (0%, 15%, 30%, and 50%) will be
placed. The pavement structure will have a 450 mm granular base, 100 mm asphalt stabilized open
graded drainage layer (ODGL), and 250 mm JPCP. Each section will have a pavement thickness of
250 mm and a pavement width of 8.5 m. The thickness of 250 mm was selected to accommodate the
cover requirements of the automatic dowel bar inserter and to reduce vibrator trails. There are
expected to be some difficulties creating longitudinal and transverse tining. The CAC feels
longitudinal tining is the most important.
Equipment
A premix plant is preferred to provide consistency in the mix. The concrete will be placed with a
double width, slip-form paver. Texturization will follow behind the paver. Dufferin will look after
organizing the equipment.
Schedule
The preliminary mix designs will require 1-2 days. About day will be required to do the final
granular base grading prior to the site survey to set the string line for the paver.
Instrumentation
The sensor cable should be placed below the granular base. Sensors should be placed below the
dowel baskets when possible. If necessary blockouts could potentially be used for sensors needing to
be placed in the top half of the pavement. Sensors will be brought to next meeting for further
discussion on placement.
174
Additional Items
The MTO representative is concerned about the feasibility of a dowel bar misalignment study in the
test sections due to the size of the test track and the cost. If there is a dowel bar misalignment study
then he would like to see the results presented in a published paper.
Action Items
The following are items that require further information:
Tim Wilson will calculate a construction estimate.
Jai Tiwari and Susan Tighe will confirm the source of aggregate.
Jai Tiwari and James Smith will develop the mix designs
John Zavarella will organize the equipment.
James Smith will plan the instrumentation.
Susan Tighe will provide information on dowel bars.
175
Meeting #3
May 4, 2007 UW/CAC/Dufferin (at Dufferin Construction)
Materials
The Steed and Evans RCA stockpile at Head Office is acceptable; however, the stockpile at the
Hiedleburg plant is not suitable for this project. Steed and Evans will sieve the RCA. The sources of
virgin aggregate are as follows: 40 mm from Dufferins Acton Pit, 19 mm form Dufferins Milton Pit,
and sand from Dufferins Forwell pit.
Design
The length of the test section is approximately 170m as measured from the guardrail to the existing
asphalt. The joint spacing will be based on the MTO specification of 3.7m, 4.5m, 4m,4.3m, and then
repeated.
Schedule
Crushing of the RCA should begin within the next two weeks. Since the best date for Dufferin is the
week of July 9, site preparation should occur the week before. The 14-day compressive strengths
should be available mid-June. The estimated date to open the slab to traffic would be July 16-17
assuming that 80% of 28-day strength is achieved. Dufferin is also available mid-August or
September.
Equipment
CoCo paving has not responded to the request to borrow their longitudinal tining kit.
Instrumentation
The strain gauge, vertical extensometer, relative humidity probe, and moisture probe were taken to
the meeting. The inter-panel extensometer was not available at the time of the meeting. Strain gauges
will be fastened to a chair structure and mound with concrete and then placed between the vibrators of
the paver. The vertical extensometer, relative humidity probes, and TDR moisture probes can be
placed below the dowel baskets. The
inter-panel extensometer will require a blockout. James Smith and John Zavarella. will provide a
design for the chair and blockouts.
Additional Work
176
The dowel bar misalignment work will not to be done since 3-4 slabs can be locked up before there
is a problem. The impact of dowel bar side shift will be looked at in the test sections since side shift
leads too horseshoe cracking. To study side shift, the dowel baskets are to be placed at the
specified intervals and the location of the saw cut will be moved to provide the side shift.
Funding
Dufferin believes the maximum tax credit they can recover is 20%. There is potential to get discounts
from suppliers if they wave their mark-up. The green municipal fund program may be an additional
source of funding. Since the concrete testing can be done inhouse at Dufferin, funds will not be
required to hire a laboratory.
Action Items
The following are items that require further information:
Tim Wilson will provide a construction estimate and letter for the University of Waterloo.
James Smith and Jai Tiwari will review the aggregate source.
James Smith and Jai Tiwari will provide the mix designs.
John Zavarella will organize the equipment.
Wayne Lazzarato and Susan Tighe will provide information for the dowel bar study.
Sources of funding will be reviewed by Wayne Lazzarato, Susan Tighe, and Rico Fung.
177
Meeting #4
May 28, 2007 UW/CAC/Dufferin (at Dufferin Construction)
Materials
The concrete sand from the Blair pit will be used in the design and is approved for structural and
pavement use. The Forwell plant is currently using St. Marys cement. St. Marys cement has a
slower initial strength gain over the first 21 days compared to Dufferins St. Lawrence cement.
Design
Preliminary testing will be started on mixes using cement content of 315, 330, and 345 kg/m3
respectively. It is projected that the desired design strength will be achieved with a cement content of
330 kg/m3 for all RCA percentages being tested. Also, the desired design strength should be
achieved with a cement content of 315 kg/m3 when combined with 0% or 15% RCA. A 25% slag
replacement by mass will be used in all mixes.
Equipment
CoCo Pavings longitudinal tining kit is available to use. CoCo Pavings mechanic will require one
day to install the tining kit on Dufferins texture/cure machine.
Schedule
The landfill requires that the test sections be open to traffic by July 1. If this window of opportunity
is missed then paving will not occur. In order to accommodate, Dufferin can push back the start of
the HWY 410 job by a week. The new schedule will be set as follows:
June 4 layout,
June 5 grade crew,
June 6 Line crew,
June 7 ODGL,
June 8 paver setup,
June 11 float day,
June 12 start paving,
June 13 change tining equip and reset for second day of paving,
June 14 paving day 2, and
June 15 float day
178
Instrumentation
The sensors have not been ordered yet and will require 3-4 weeks. The CAC will look into getting a
maturity sensor from The Transtec Group. A request for a rush on the sensors from Hoskin/RST will
be made. Also, availability of sensors through other manufactures (Slope Indicator, CTL Group)
will be explored in order to meet the tight time frame.
Funding
Steed and Evans are donating the aggregate and screening. CoCo paving is donating the longitudinal
tine kit. Dufferin is discounting the price of the equipment, materials and labour to the actual cost
price, which is estimated at $185,000. If Dufferin receives a 20% research and development credit
for the cost of this project (not including labour) then the University of Waterloo will have a
corresponding reduction in their bill.
179
Meeting #5
May 31, 2007 UW/CAC/Dufferin (Conference Call)
Contacts
Dufferin Site Supervisor is Jerry Armstrong (416-984-4480).
Landfill Site Supervisor is Dave McCaughan (519-897-3009). Note: the site is locked every night.
Materials and Design

The 12 trial mixes have been batched, cylinders demolded and are curing. The 7-day compressive
strengths will be available next week. Stantec completed falling weight deflectometer (FWD) testing
on the existing 450mm granular base and the results exceeded the estimated strength requirements in
the design. The test track area where the 30% RCA mix will be placed had a weaker FWD result
than other areas of the test track; however, it still exceeded the estimated strength requirements.
St. Marys cement will reach or exceed the design strength at 28 days as long as 65-70% of the
compressive strength is reached by 21 days.
Construction
The distance from the guardrail to the existing asphalt has been confirmed at 172m. Dufferin will
remove approximately 8m of the existing asphalt to allow for a transition zone. Construction Testing
Asphalt Lab Ltd. Has completed the asphalt stabilized OGDL mix design. A 2% crossfall is required
on the pavement from the centreline. The concrete haul time from the Forwell plant to the test track
is approximately 20 minutes.
Equipment
The fresh concrete will be transported in ready mix trucks since dump trucks are unavailable. The
longitudinal tines will be about 0.5m shorter than the full pavement width.
Schedule
The schedule is unchanged from the last meeting. There is a potential to move the second day of
paving to Wednesday depending on the success of the first day.
Instrumentation
180
The maturity sensors have been shipped by the Transtec Group and will arrive in the next couple of
days. Hoskin/RST will be able to deliver the strain gauges, vertical extensometer, and inter-panel
extensometer in time for construction. The relative humidity and TDR moisture probes cannot be
delivered before construction begins and will not be retrofitted at a later time.
Additional Work
The dowel bar side shift study parameters will be confirmed with the MTO.
181
Meeting #6
June 11, 2007 Onsite Construction
Initially there was a quick site tour and an area for the concrete truck washout was identified.
Construction
All tie dowel baskets are to be cut to improve MIT scan results. The pins to secure the load transfer
baskets are anchored at every second dowel on both sides.
The section lengths were revised based on a total area of 180m (0% - 30m, 15% - 50m, 30% - 50m,
50% - 50m). These lengths are estimates, as the paver will continue to move forward so there will be
no hard edge.
The concrete will arrive at the test track with 40mm slump.
The dowel bar side shift distances of 0, 25, 50, 75, 100mm will be tested. The MTO rejects dowel bar
side shift distances greater than 50mm. A construction drawing will be made by the University of
Waterloo to assist in the dowel bar placement. Dowel bar offsets will be placed as follows: 0mm
offset for the first 10 joints, 25mm offset for the next 6 joints, 50mm offset for the next 6 joints,
75mm offset for next 6 joints, 100mm offset for next 6 joints, and 0mm offset for the remaining 8
joints.
Instrumentation
The maturity sensors from the Transtec Group and the strain, vertical extensometer, inter-panel
extensometer, and maturity sensors from Hoskin/RST have arrived. The datalogger and remote
multiplexers will arrive about 2-3 weeks after construction and will be installed at that time.
182
Appendix C
Pavement Distress
Ravelling and Coarse Aggregate Loss

Very Slight Barley noticeable.
Slight Noticeable loss of pavement material.
Moderate Pavement has pockmarked appearance, with pockmarks fairly well spaced. Shallow
disintegration of pavement surface. An open-texture look.
Severe Pavement has pockmarked appearance, with pockmarks closely spaced.
Disintegration with small, shallow potholes.
Very Severe Surface has a raveled and disintegrated appearance with large, shallow potholes.
Polishing
Very Slight Barely noticeable.
Slight Noticeable dull finish.
Moderate Distinctive dull finish.
Severe Glossy mirror finish.
Very Severe Surface has a highly polished appearance.
Scaling
Very Slight Barely noticeable.
Slight Noticeable.
Moderate An open-texture look, as with raveling, by very shallow.
Severe Disintegration in closely spaced, shallow patched.
Very Severe Disintegration in shallow large patched.
183
Potholing
Very Slight Barely noticeable. Pothole resembles a pop-out of coarse aggregate.
Slight Disintegration of surrounding materials.
Moderate Pothole much wider (<75 mm) than a pop-out of coarse aggregate and deeper (< 75
mm)
Severe Pothole 75 150 mm wide and 75 150 mm deep.
Very Severe Pothole over 150 mm winde and over 150 mm deep. Interferes with rideability.
Joint and Cracking Spalling

Very Slight Small crack(s) with very small fractures.
Slight Small crack(s) within 75 mm of the joint or crack, with a few small pieces missing or
loosened from the fracture area.
Moderate Spalling extends more than 75 mm of the joint or crack, with many small pieces
missing or loosened from the fracture area.
Severe Spalling extends more than 75 mm of the joint or crack, with larger pieces missing or
loosened from the fracture area. Temporary patching may have been placed,
Very Severe Large potholes are at the places along the joint or crack, perhaps causing tire damage.
Faulting
Very Slight Barely noticeable (<3 mm).
Slight 3 6 mm.
Moderate 7 12 mm.
Severe 13 19 mm.
Very Severe > 19 mm.
184
Distortion
Very Slight Barely noticeable swaying of vehicle.
Slight Barely noticeable pitch and role, and a jarring bump or drop of vehicle.
Moderate Noticeable pitch and role, and a harsh bump or jarring of vehicle.
Severe A continuous pitch and roll, and a hard jarring bump or drop of vehicle. The driver
always must anticipate distortion ahead.
Very Severe Continuous distortion, making the driver feel it necessary to reduce spped from the
posted speed limit.
Sealant Loss
Very Slight Barely popped out or breaking.
Slight Sealant broken and beginning to pull out (up to 30 cm)
Moderate Sealant broken and pulled out by up to 50% of its length.
Severe Sealant broken and pulled out by up to 58% of its length.
Very Severe Sealant is completely broken and pulled out by more than 80% of its length. It is
ineffective as a sealant.
Joint Failure
Severe Pavement fractures into blocks, with multiple cracks and missing pieces along both
sides of the joint. Distortion is noticeable.
Very Severe Pavement fractures into large blocks, with multiple cracks and missing pieces along
both sides of the joint, extending a considerable distance (2 3 m ) from the joint.
Distortion is noticeable.
Longitudinal and Meandering Cracking

Very Slight < 3 mm wide.
Slight 3 12 mm wide.
Moderate 13 19 mm wide (with or without spalling and faulting).
Severe 20 25 mm wide (with spalling and faulting).
Very Severe > 25 mm (with spalling and faulting).
185
Diagonal, Corner, and Edge Crescent
Very Severe > 25 mm wide (with spalling and faulting).
D Cracking
Moderate 13 19 mm wide, or multiple cracks < 12 mm wide.
Severe 20 25 mm wide, or multiple cracks 13 19 mm wide.
Very Severe > 25 mm wide, or multiple cracks 20 25 mm wide.
Transverse Cracking
Very Severe > 25 mm wide (with spalling and faulting).
186
Appendix D
Datalogger Program
'CR1000 Series Datalogger
'Program name: WO Q08619 4 remote mux
'Date written: 22/06/07 DWP RST Inst
'Modifications: corrected Sub convert (matching serial number and channel) and increasing mux delay to 15 (to
account for longer cable lengths)
'\\\\\\\\\\\\\\\\\\\\\\\\\ DECLARATIONS /////////////////////////
Public VWSG(24), VWPAVE(8), VWEXT(8)
Public SGTH(24), PAVETH(8), EXTTH(8)
Public Pave_Disp(8), Ext_Disp(8)
Public Bat_V
Public Int_Temp
Public Flag(4) as boolean
public i1
Units VWSG()=uE
Units Ext_Disp()=mm
Units Pave_Disp()=mm
'\\\\\\\\\\\\\\\\\\\\\\\\ OUTPUT SECTION ////////////////////////
DataTable(DATA,-1,-1)' >>
CardOut (0 ,-1)
Sample(24,VWSG(),IEEE4)
Sample(8,Pave_Disp(),IEEE4)
Sample(8,Ext_Disp(),IEEE4)
Sample(8,VWPAVE(),IEEE4)
Sample(8,VWEXT(),IEEE4)
Sample(24,SGTH(),FP2)
Sample(8,PAVETH(),FP2)
Sample(8,EXTTH(),FP2)
187
Sample(1, Bat_V, FP2)
Sample(1, Int_Temp, FP2)
EndTable
Sub VW_SG
Delay(0,5,MSEC)
VibratingWire(VWSG(i1),1,mV7_5,1,Vx1,400,1200,250,-1,20000,100,0,3405,0) ':read
Vibrating Wire Piezometer
BRHalf(SGTH(i1), 1, mV2500, 2, VX1, 1, 2500, False, 10000, 250, 2.5, 0) ':read 3K Thermistors
SGTH(i1)=-104.78+378.11*SGTH(i1)+-611.59*SGTH(i1)^2+544.27*SGTH(i1)^3+-
240.91*SGTH(i1)^4+43.089*SGTH(i1)^5
':linearize 3K thermistors
EndSub
Sub VW_PAVE
Delay(0,5,MSEC)
VibratingWire(VWPAVE(i1),1,mV7_5,1,Vx1,1600,2800,200,-1,20000,150,0,1,0) ':read
Vibrating Wire Piezometer
BRHalf(PAVETH(i1), 1, mV2500, 2, VX1, 1, 2500, False, 10000, 250, 2.5, 0) ':read 3K thermistors
PAVETH(i1)=-104.78+378.11*PAVETH(i1)+-611.59*PAVETH(i1)^2+544.27*PAVETH(i1)^3+-
240.91*PAVETH(i1)^4+43.089*PAVETH(i1)^5
EndSub
Sub VW_EXT
Delay(0,5,MSEC)
VibratingWire(VWEXT(i1),1,mV7_5,1,Vx1,1800,3200,200,-1,20000,200,0,1,0) ':read Vibrating Wire

Piezometer
BRHalf(EXTTH(i1), 1, mV2500, 2, VX1, 1, 2500, False, 10000, 250, 2.5, 0) ':read 3K thermistors
EXTTH(i1)=-104.78+378.11*EXTTH(i1)+-611.59*EXTTH(i1)^2+544.27*EXTTH(i1)^3+-
240.91*EXTTH(i1)^4+43.089*EXTTH(i1)^5
188
EndSub
Sub convert
'convert to displacement - serial number must match channel
Pave_Disp(1)=(2.416-VWPAVE(1))*2.932' ENTER Pavement VM0220 CAL FACTORS *
Pave_Disp(5)=(2.639-VWPAVE(5))*2.936' ENTER Pavement VM0222 CAL FACTORS
Ext_Disp(1)=(2.692-VWEXT(1))*2.973' ENTER INTERPANEL VM0228 CAL FACTORS *
Ext_Disp(5)=(2.838-VWEXT(5))*2.968' ENTER INTERPANEL VM0230 CAL FACTORS
' m b
'218 0.0029760 2698.9
'219 0.0029493 2666.6
'220 0.0029316 2415.5
'221 0.0029900 2503.7
'222 0.0029360 2639.3
'223 0.0029738 2678.3
'224 0.0029930 3137.9
189
'225 0.0030103 2699.2
'226 0.0029722 2478.6
'227 0.0029836 2749.9
'228 0.0029726 2692.4
'229 0.0029399 3023.9
'230 0.0029684 2838.0
'231 0.0029540 2742.7
'232 0.0029387 3223.1
'233 0.0029619 2759.2
EndSub
'\\\\\\\\\\\\\\\\\\\\\\\\\\\ PROGRAM ////////////////////////////
BeginProg
Scan(2,hr, 3, 0) ' >200 days @ 1 record / 2 hr
PortSet(1, 1) ':ENABLE REMOTE Multiplexer 1 & 2
Delay(0,15,MSEC)
For i1 = 1 to 6 'REMOTE MUX1 SG 1-6
PulsePort(2,5000) 'CLOCK MUX0,5,MSEC)
Call VW_SG
Next i1
For i1 = 1 to 2 'MUX1 PAVE 1-2
Call VW_PAVE
Next i1
For i1 = 1 to 2 'MUX1 EXT 1-2
Call VW_EXT
Next i1
190
Call VW_SG
Next i1
Call VW_PAVE
Next i1
Call VW_EXT
Next i1
PortSet(1, 0) ':TURN OFF THE REMOTE MUX 1 & 2
PortSet(3, 1) ':ENABLE REMOTE Multiplexer 3 & 4
Delay(0,15,MSEC)
Call VW_SG
Next i1
Call VW_PAVE
Next i1
Call VW_EXT
191
Next i1
Call VW_SG
Next i1
Call VW_PAVE
Next i1
Call VW_EXT
Next i1
PortSet(3, 0) ':TURN OFF THE REMOTE MUX 1 & 2
Battery(Bat_V)
PanelTemp(Int_Temp, 250)
Call convert
CallTable DATA
NextScan
EndProg
192
Appendix E
Sensor Results
0% Coarse RCA Strain Sensor #7
193
194
195
196
197
198
199
200
0% Coarse RCA Vertical Extensometer #3
201
202
203
0% Coarse RCA Inter-Panel Extensometer #3
204
205
206

PHD Thesis Jsmith

Uploaded by

Copyright:

Available Formats

PHD Thesis Jsmith

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

PHD Thesis Jsmith

Uploaded by

Copyright:

Available Formats

Recycled Concrete Aggregate A Viable

Aggregate Source For Concrete Pavements

James Trevor Smith

Waterloo, Ontario, Canada, 2009

James Trevor Smith 2009

1.1.1 Current Aggregate Shortage

1.1.2 Impacts of Waste Concrete

1.1.4 Achieving Sustainability with Recycled Aggregate Concrete

1.1.5 Barriers to Recycled Concrete Aggregate Use

1.2 Thesis Scope and Objectives

This research has the following objectives.

1.3 Project Significance

Sustainability is defined by the World Commission on Environment and Development as Meeting

2.2 Recycled Concrete Aggregate

2.2.1 Producing Recycled Concrete Aggregate

Figure 2.1 shows a closed-loop concrete system [Kuroda 2005].

Figure 2.2 Cone and Jaw Compression Crushers

Figure 2.3 Vertical and Horizontal Impact Crushers

2.2.2 Current Use of Recycled Concrete Aggregate

Figure 2.4 United States Using RCA as Base Material

Figure 2.5 United States Using RCA as Concrete Aggregate

2.2.3 Properties of RCA

2.2.3.1 Shape, Texture, and Gradation

2.2.3.2 Specific Gravity

2.2.3.4 Abrasion Resistance

2.2.3.5 RCA Fines

2.2.4 Properties of RCA Concrete

2.2.4.3 Air Content

2.2.4.4 Initial Set Time

2.2.4.5 Final Set Time

2.2.4.7 Flexural Strength

f r' = 0.94 f c' (2.1)

where f r' = flexural strength (MPa)

2.2.4.11 Modulus of Elasticity

E c = 0.043W c1.5 f c' (2.2)

2.2.4.13 Dry Shrinkage

2.2.4.14 Alkali-Aggregate Reactivity

2.2.4.16 Depth of Carbonation

2.2.4.17 Sulfate Resistance

2.2.4.19 Chloride Penetration

2.2.4.20 Chloride Conductivity

2.2.4.21 Water Absorption

2.2.4.23 Skid Resistance

2.2.4.24 Ultra Sound Pulse Velocity

2.2.4.25 Coefficient of Thermal Expansion

2.2.4.26 Pore Structure

2.2.5 RCA Properties Summary

Table 2.2 RCA Concrete Properties Summary

2.3 RCA Concrete Mix Designs

2.3.1 Components of concrete mixes

2.3.2 Mixing Methods

2.3.3.1 Aggregate Testing

Specific gravity (kg/m3) > 2.2 > 2.5 > 2.5

2.3.3.2 Concrete Testing

Table 2.4 Maximum Limit of Harmful Substances Allowed in RCA Concrete

Table 2.5 RCA Mix Designs

2.4.1 FHWA Recycled Materials Policy

2.4.2 LEED Program