The Prediction of Coarse Aggregate Performance by Micro-Deval and Soundness Related Aggregate Tests

Publications Page

Table of Contents

RESEARCH REPORT ICAR 507-1F

Sponsored by the Aggregates Foundation for Technology, Research and Education

1. Report No. ICAR 507-1F 4. Title and Subtitle

2. Government Accession No.

3. Recipient's Catalog No. 5. Report Date June 2006 6. Performing Organization Code

The Prediction of Coarse Aggregate Performance by Micro-Deval and Other Aggregate Tests
7. Author(s) Dr. David W. Fowler, Dr. John J. Allen, Alexander Lange, and Peter Range 9. Performing Organization Name and Address International Center for Aggregates Research The University of Texas at Austin 4030 W Braker Lane, Bldg. 200, Ste 252 Austin, Texas 78759-5329 12. Sponsoring Agency Name and Address Aggregates Foundation for Technology, Research, and Education 1605 King Street Alexandria, VA 22314 15. Supplementary Notes

8. Performing Organization Report No. Research Report ICAR 507-1F 10. Work Unit No. (TRAIS) 11. Contract or Grant No. Project No. 507 13. Type of Report and Period Covered September 2003 August 2006 14. Sponsoring Agency Code

16. Abstract

This research project concentrated on determining whether or not a correlation existed between laboratory aggregate tests and observed aggregate field performance. For this purpose, aggregate samples were collected from the majority of the U.S. states as well as several Canadian provinces and subjected to a variety of strength, soundness, and intrinsic particle property tests. Additionally, performance data on the aggregates was obtained by contacting multiple DOTs where aggregates were in use in several categories hot-mix asphalt, portland cement concrete, base course, and open-graded friction course. Numerical and qualitative analyses were performed to evaluate the success of separating good performers from fair and poor performers using the micro-Deval test alone as well as the micro-Deval test combined with another test. Special attention was paid to aggregate mineralogical composition. Furthermore, attempts were made to determine if a correlation exists between any two tests.

17. Key Words Aggregates, hot-mix asphalt, portland cement concrete, base course, open-graded friction course, micro-Deval test, aggregate minerlogical composition.

18. Distribution Statement No restrictions.

19. Security Classif.(of this report) Unclassified

Technical Report Documentation Page

20. Security Classif.(of this page) Unclassified

21. No. of Pages

22. Price

Form DOT F 1700.7 (8-72)

Reproduction of completed page authorized

The Prediction of Coarse Aggregate Performance by Micro-Deval and Other Aggregate Tests

Dr. David W. Fowler Dr. John J. Allen Alexander Lange, and Peter Range The University of Texas at Austin Austin, Texas

ICAR Report 507-1F Application and Significance of the Micro-Deval Test

Sponsored by: International Center for Aggregates Research The University of Texas at Austin Aggregates Foundation for Technology, Research and Education (AFTRE)

July, 2006
ii

Abstract
This research project concentrated on determining whether or not a correlation existed between laboratory aggregate tests and observed aggregate field performance. For this purpose, aggregate samples were collected from the majority of the U.S. states as well as several Canadian provinces and subjected to a variety of strength, soundness, and intrinsic particle property tests. Additionally, performance data on the aggregates was obtained by contacting multiple DOTs where aggregates were in use in several categories hot-mix asphalt, portland cement concrete, base course, and open-graded friction course. Numerical and qualitative analyses were performed to evaluate the success of separating good performers from fair and poor performers using the micro-Deval test alone as well as the micro-Deval test combined with another test. Special attention was paid to aggregate mineralogical composition. Furthermore, attempts were made to

determine if a correlation exists between any two tests.

iii

Table of Contents
List of Tables ........................................................................................................ xii List of Figures ........................................................................................................xv Chapter 1: Introduction ...........................................................................................1 1.1 1.2 1.3 Need for Project .....................................................................................1 Project Objective....................................................................................1 Scope of Project .....................................................................................2

Chapter 2: Review of Literature .............................................................................3 2.1 2.2 2.3 Introduction............................................................................................3 Field Performance Criteria.....................................................................3 Micro-Deval...........................................................................................5 2.3.1 Background ...................................................................................5 2.3.2 Significance...................................................................................5 2.3.3 Correlations with Other Tests .......................................................8 2.3.4 Precision and Variables...............................................................11 2.3.5 Current Use .................................................................................12 2.4 Los Angeles Abrasion..........................................................................13 2.4.1 Background .................................................................................13 2.4.2 Significance.................................................................................14 2.4.3 Precision......................................................................................17 2.4.4 Current Use .................................................................................17 2.5 Other Abrasion Tests ...........................................................................18

2.6

Soundness Tests ...................................................................................20 2.6.1 Background .................................................................................20 2.6.2 Significance.................................................................................21 2.6.4 Current Use .................................................................................25

2.7

Freezing and Thawing..........................................................................25 2.7.1 Background .................................................................................25 2.7.2 Significance.................................................................................27 2.7.3 Current Use .................................................................................29

2.8

Petrographic Evaluation.......................................................................30 2.8.1 Background .................................................................................30 2.8.2 Significance.................................................................................33 2.8.3 Precision......................................................................................34 2.8.4 Current Use .................................................................................34

2.9

Strength and Impact .............................................................................35 2.9.1 Significance.................................................................................35 2.9.2 Current Use .................................................................................37

2.10 Absorption............................................................................................37 2.10.1 Significance...............................................................................37 2.10.2 Current Use ...............................................................................38 2.11 Aggregate Shape and Surface Texture.................................................39 2.11.1 Significance...............................................................................39 2.11.2 Current Use ...............................................................................39 Chapter 3: Aggregate Acquisition and Preparation ..............................................41 3.1 3.2 3.3 3.4 3.5 Introduction..........................................................................................41 Initial Survey........................................................................................41 Aggregate Test Suite Determination....................................................42 Field Performance Rating Determination ............................................44 Aggregate Acquisition .........................................................................47

vi

Chapter 4: Field Performance Determination .......................................................51 4.1 4.2 Introduction..........................................................................................51 Rating System ......................................................................................51

Chapter 5: Testing.................................................................................................55 5.1 5.2 Introduction..........................................................................................55 Aggregate Sample Preparation ............................................................55 5.2.1 Processing ...................................................................................55 5.2.2 Standardizing Gradations............................................................56 5.2.3 Performing Test Procedures........................................................57 5.3 Petrographic Analysis ..........................................................................59 5.3.1 Test Procedures...........................................................................59 5.3.2 Results.........................................................................................59 5.4 Micro-Deval.........................................................................................59 5.4.1 Test Procedures...........................................................................59 5.4.2 Results.........................................................................................61 5.5 Magnesium Sulfate Soundness ............................................................62 5.5.1 Test Procedures...........................................................................62 5.5.2 Results.........................................................................................64 5.5.3 Lab Data Comparison .................................................................66 5.6 Los Angeles Abrasion..........................................................................67 5.6.1 Test Procedures...........................................................................67 5.6.2 Results.........................................................................................68 5.7 Freezing and Thawing..........................................................................69 5.7.1 Test Procedures...........................................................................69 5.7.2 Results.........................................................................................73 5.8 Aggregate Crushing Value Test...........................................................74 5.8.1 Test Procedures...........................................................................74 5.8.2 Results.........................................................................................75

vii

5.9

Aggregate Crushing Value (SSD) Test................................................75 5.9.1 Test Procedures...........................................................................75 5.9.2 Results.........................................................................................76

5.10 Absorption Test....................................................................................77 5.10.1 Test Procedures.........................................................................78 5.10.2 Results.......................................................................................79 5.11 Specific Gravity Test ...........................................................................79 5.11.1 Test Procedures.........................................................................79 5.11.2 Results.......................................................................................80 5.12 Flat and Elongated Test .......................................................................80 5.12.1 Test Procedures.........................................................................81 5.12.2 Results.......................................................................................83 5.13 Percent Fractured Particles ..................................................................83 5.13.1 Test Procedures.........................................................................83 5.13.2 Results.......................................................................................83 Chapter 6: Discussion and Analysis .....................................................................85 6.1 Introduction..........................................................................................85 6.1.1 Analysis Methodology ................................................................85 6.1.2 Rock Type Subgroups.................................................................87 6.1.3 Climatic Regions Subgroups.......................................................87 6.2 Performance Analysis for Hot-Mix Asphalt Aggregates.....................89 6.2.1 Individual Tests...........................................................................89 6.2.2 Combinations Involving Micro-Deval......................................105 6.2.3 Other Relevant Combinations...................................................118 6.2.4 Results Summary ......................................................................121 6.2.5 Limestone and Dolomite...........................................................122 6.2.5.1 Combinations Involving Micro-Deval..........................122 6.2.5.2 Other Relevant Combinations.......................................131 6.2.5.3 Section Results Summary .............................................132

viii

6.2.6 Siliceous Gravel ........................................................................133 6.2.6.1 Combination Involving Micro-Deval ...........................133 6.2.6.2 Other Relevant Combinations.......................................143 6.2.6.3 Section Results Summary .............................................144 6.2.7 Granite.......................................................................................145 6.2.7.1 Combinations Involving Micro-Deval..........................145 6.2.7.2 Other Relevant Combinations.......................................154 6.2.7.3 Section Results Summary .............................................155 6.2.8 Climatic Regions.......................................................................156 6.2.8.1 Region II .......................................................................156 6.2.8.2 Region III ......................................................................159 6.2.8.3 Region V .......................................................................163 6.2.8.4 Region VI......................................................................165 6.3 Performance Analysis for Portland Cement Concrete Aggregates....167 6.3.1 Individual Tests.........................................................................167 6.3.2 Combinations Involving Micro-Deval......................................177 6.3.3 Other Relevant Combinations...................................................186 6.3.4 Results Summary ......................................................................187 6.3.5 Limestone and Dolomite...........................................................188 6.3.5.1 Combinations Involving Micro-Deval..........................188 6.3.5.2 Other Relevant Combinations.......................................193 6.3.5.3 Section Results Summary .............................................194 6.3.6 Siliceous Gravel ........................................................................194 6.3.6.1 Combinations Involving Micro-Deval..........................194 6.3.6.2 Other Relevant Combinations.......................................199 6.3.6.3 Section Results Summary .............................................200 6.4 Performance Analysis for Base Course Aggregates ..........................201 6.4.1 Individual Tests.........................................................................201 6.4.2 Combinations Involving Micro-Deval......................................201 6.4.3 Other Relevant Combinations...................................................212 6.4.4 Results Summary ......................................................................213 ix

6.5

Performance Analysis for Open-Graded Friction Course Aggregates214 6.5.1 Individual Tests.........................................................................214 6.5.2 Combinations Involving Micro-Deval......................................223 6.5.3 Other Relevant Combinations...................................................232 6.5.4 Results Summary ......................................................................235

6.6

Test Correlations Analysis.................................................................236 6.6.1 Correlations Analysis Overview ...............................................236 6.6.2 Complete Data Set Regression Analysis...................................238 6.6.3 Partial Data Set Regression Analysis I .....................................245 6.6.4 Partial Data Set Regression Analysis II ....................................246 6.6.5 Results Summary ......................................................................248

6.7

Micro-Deval and AIMS Analysis ......................................................249 6.7.1 Background Information...........................................................249 6.7.2 Testing Procedure .....................................................................250 6.7.3 Results Discussion ....................................................................251 6.7.4 Summary and Conclusions .......................................................256

Chapter 7: Summary and Conclusions................................................................257 7.1 7.2 7.3 Project Summary................................................................................257 Major Conclusions .............................................................................258 Need for Future Research ..................................................................259

Appendix A: Performance Questionnaire...........................................................261 Appendix B: Micro-Deval Test Results..............................................................265 Appendix C: Magnesium Sulfate Soundness Test Results .................................269 Appendix D: Los Angeles Abrasion Test Results ..............................................273 Appendix E: Canadian Freeze-Thaw Test Results .............................................277 Appendix F: Aggregate Crushing Value Test Results........................................281 Appendix G: Aggregate Crushing Value (SSD) Test Results ............................285 Appendix H: Absorption and Specific Gravity Test Results ..............................289 Appendix I: Particle Shape Factor Test Results .................................................295 Appendix J: Percent Fractured Particles Test Results ........................................301 Appendix K: Petrographic Analysis ...................................................................305 Appendix L: Performance Data ..........................................................................309 Appendix M: Hot-Mix Asphalt Aggregates Graphs...........................................315 Appendix N: Portland Cement Concrete Aggregates Graphs.............................353 Appendix O: Base Course Aggregates Graphs...................................................401 Appendix P: Open-Graded Friction Course Aggregates Graphs........................439 Appendix Q: Test Correlation Graphs for the Full Data Set ..............................486 Appendix R: Test Correlation Graphs for the Partial Data Set I ........................515 Appendix S: Test Correlation Graphs for the Partial Data Set II .......................545 Appendix T:Test Correlation Tables for Full Data Set, Partial Data Set I, and Partial Data Set II 575 Appendix U: Dr. Eyad Masad AIMS Report......................................................586 References............................................................................................................601

xi

List of Tables
Table 2.2-1: Table 2.2-2: Table 3.4-1: Table 3.5-1: Table 3.5-2: Table 4.2-1: Table 5.4.1-1: Table 5.5.1-1: Table 5.5.3-1: Table 6.1.1-1: Table 6.2.1-1: Table 6.2.1-2: Table 6.2.1-3: Table 6.2.1-4: Table 6.2.1-5: Table 6.2.1-6: Table 6.2.1-7: Table 6.2.1-8: Table 6.2.1-9: Table 6.2.1-10: Table 6.2.1-11: Table 6.2.2-1: Table 6.2.2-2: Table 6.2.2-3: Table 6.2.2-4: Table 6.2.2-5: Table 6.2.2-6: Table 6.2.2-7: Table 6.2.2-8: Table 6.2.3-1: Table 6.2.3-2: Table 6.2.4-1: Table 6.2.5.1-1: Table 6.2.5.1-2: Table 6.2.5.1-3: Table 6.2.5.1-4: Table 6.2.5.1-5: Table 6.2.5.3-1: Table 6.2.6.1-1: Table 6.2.6.3-1: Table 6.2.7.1-1: Classification System Used by Rogers and Senior .................................... 4 Performance Evaluation Used by Wu et al. ............................................... 4 ICAR 507 Evaluation Criteria for Determining the Performance Rating of Aggregates ............................................................................... 46 Aggregate Qualifications per Application ............................................... 48 Division of Aggregates by Geological Rock Type .................................. 50 Performance Criteria Developed for Use in This Project ........................ 52 Gradations for Use in Micro-Deval Testing ............................................ 60 Sulfate Soundness Sample Gradations..................................................... 62 MSS Test Data and Procedures Summary ............................................... 66 Analysis Formula Definitions .................................................................. 86 Micro-Deval Success Rate....................................................................... 90 MSS Success Rate.................................................................................... 92 LAA Success Rate ................................................................................... 93 CFT Success Rate .................................................................................... 94 ACV Success Rate ................................................................................... 96 WCV Success Rate .................................................................................. 97 Absorption Success Rate.......................................................................... 99 SG (Bulk) Success Rate ......................................................................... 100 PSF Success Rate................................................................................... 102 Percnt Crushed (1+) Success Rate ......................................................... 103 Percent Crushed (2+) Success Rate ....................................................... 105 LAA and MD Success Rate ................................................................... 107 MSS and MD Success Rate ................................................................... 108 CFT and MD Success Rate .................................................................... 110 ACV and MD Success Rate................................................................... 112 WCV and MD Success Rate .................................................................. 113 ABS and MD Success Rate.................................................................... 114 SG(Bulk) and MD Success Rate............................................................ 116 PSF and MD Success Rate..................................................................... 117 ACV and CFT Success Rate .................................................................. 119 PSF and MSS Success Rate ................................................................... 120 HMA Success Rates Summary .............................................................. 121 MD Success Rate ................................................................................... 123 MSS vs. MD Success Rate..................................................................... 124 LAA vs. MD Success Rate .................................................................... 125 CFT vs. MD Success Rate ..................................................................... 126 ACV vs. MD Success Rate .................................................................... 127 HMA Limestone and Dolomite Success Rate Summary....................... 133 MD Success Rate ................................................................................... 134 HMA Siliceous Gravel Success Rate Summary .................................... 145 MD Success Rate ................................................................................... 146 xii

Table 6.2.7.3-1: Table 6.2.8.1-1: Table 6.2.8.1-2: Table 6.2.8.1-3: Table 6.2.8.1-4: Table 6.2.8.2-1: Table 6.2.8.2-2: Table 6.2.8.2-3: Table 6.2.8.2-4: Table 6.2.8.2-5: Table 6.2.8.3-1: Table 6.2.8.3-2: Table 6.2.8.4-1: Table 6.2.8.4-2: Table 6.3.1-1: Table 6.3.1-2: Table 6.3.1-3: Table 6.3.1-4: Table 6.3.1-5: Table 6.3.1-6: Table 6.3.1-7: Table 6.3.1-8: Table 6.3.1-9: Table 6.3.1-10: Table 6.3.2-1: Table 6.3.2-2: Table 6.3.2-3: Table 6.3.4-1: Table 6.3.5.1-1: Table 6.3.5.1-2: Table 6.3.5.1-3: Table 6.3.5.1-4: Table 6.3.5.1-5: Table 6.3.5.2-1: Table 6.3.5.3-1: Table 6.3.6.1-1: Table 6.3.6.1-2: Table 6.3.6.1-3: Table 6.3.6.1-4: Table 6.3.6.1-5: Table 6.3.6.2-1: Table 6.3.6.3-1: Table 6.4.2-1: Table 6.4.2-2: Table 6.4.2-3:

HMA Granite Success Rate Summary................................................... 155 MD Success Rate ................................................................................... 156 HMA Region II Success Rate Summary ............................................... 157 MD Success Rate ................................................................................... 157 HMA Region II No Limestone/Dolomite Success Rate Summary ....... 158 MD Success Rate ................................................................................... 159 LAA vs. MD Success Rate .................................................................... 160 ACV vs. MD Success Rate .................................................................... 161 WCV vs. MD Success Rate ................................................................... 162 HMA Region III Success Rate Summary .............................................. 163 MD Success Rate ................................................................................... 163 HMA Region V Success Rate Summary ............................................... 164 MD Success Rate ................................................................................... 165 HMA Region VI Success Rate Summary.............................................. 166 MD Success Rate ................................................................................... 168 MSS Success Rate.................................................................................. 169 LAA Success Rate ................................................................................. 170 CFT Success Rate .................................................................................. 171 ACV Success Rate ................................................................................. 172 WCV Success Rate ................................................................................ 173 ABS Success Rate.................................................................................. 174 SG(Bulk) Success Rate .......................................................................... 175 PSF Success Rate................................................................................... 176 Fractured Particles (2 or more sides) vs. Performance .......................... 177 MSS vs. MD Success Rate..................................................................... 177 CFT vs. MD Success Rate ..................................................................... 179 ABS vs. MD Success Rate..................................................................... 182 PCC Success Rate Summary ................................................................. 187 MD Success Rate ................................................................................... 188 MSS vs. MD Success Rate..................................................................... 189 CFT vs. MD Success Rate ..................................................................... 190 ABS vs. MD Success Rate..................................................................... 191 SG(Bulk) vs. MD Success Rate ............................................................. 192 ABS vs. CFT Success Rate.................................................................... 193 PCC Limestone and Dolomite Success Rate Summary......................... 194 MD Success Rate ................................................................................... 195 MSS vs. MD Success Rate..................................................................... 195 CFT vs. MD Success Rate ..................................................................... 196 ABS vs. MD Success Rate..................................................................... 197 SG(Bulk) vs. MD Success Rate ............................................................. 198 ABS vs. CFT Success Rate.................................................................... 199 PCC Siliceous Gravel Success Rate Summary...................................... 200 MSS and MD Success Rate ................................................................... 203 LAA and MD Success Rate ................................................................... 204 CFT and MD Success Rate .................................................................... 205 xiii

Table 6.4.2-4: Table 6.4.2-5: Table 6.4.2-6: Table 6.4.2-7: Table 6.4.2-8: Table 6.4.4-1: Table 6.5.1-1: Table 6.5.1-2: Table 6.5.1-3: Table 6.5.1-4: Table 6.5.1-5: Table 6.5.1-6: Table 6.5.1-7: Table 6.5.1-8: Table 6.5.1-9: Table 6.5.1-10: Table 6.5.2-1: Table 6.5.2-2: Table 6.5.2-3: Table 6.5.2-4: Table 6.5.2-5: Table 6.5.2-6: Table 6.5.3-1: Table 6.5.3-2: Table 6.5.3-3: Table 6.5.4-1:

ACV and MD Success Rate................................................................... 207 WCV and MD Success Rate .................................................................. 208 ABS and MD Success Rate.................................................................... 209 SG (Bulk) and MD Success Rate........................................................... 210 PSF and MD Success Rate..................................................................... 212 BC Success Rates Summary .................................................................. 213 MD Success Rate ................................................................................... 215 MSS Success Rate.................................................................................. 216 LAA Success Rate ................................................................................. 217 CFT Success Rate .................................................................................. 218 ACV Success Rate ................................................................................. 219 WCV Success Rate ................................................................................ 220 ABS Success Rate.................................................................................. 220 SG(Bulk) Success Rate .......................................................................... 221 PSF Success Rate................................................................................... 222 Fractured Particles (2 or more sides) vs. Performance .......................... 223 LAA vs. MD Success Rate .................................................................... 224 ACV vs. MD Success Rate .................................................................... 226 WCV vs. MD Success Rate ................................................................... 227 ABS vs. MD Success Rate..................................................................... 228 SG(Bulk) vs. MD Success Rate ............................................................. 229 PSF vs. MD Success Rate...................................................................... 230 ACV vs. LAA Success Rate .................................................................. 232 WCV vs. CFT Success Rate .................................................................. 233 ABS vs. ACV Success Rate................................................................... 234 OGFC Success Rate Summary .............................................................. 236

xiv

List of Figures
Figure 2.3.2-1: Example of Plot by Senior and Rogers ...................................................... 7 Figure 2.3.3-1: Correlation Between Magnesium Sulfate Soundness and Micro-Deval as Found by Senior and Rogers.......................................... 10 Figure 2.3.5-1: Correlation Between Wet and Soaked Micro-Deval Loss as Found by Woodside and Woodward........................................................ 13 Figure 2.4.1-1: Correlation Between Aggregate Impact Value and Los Angeles Abrasion as Found by Senior and Rogers................................................ 14 Figure 3.5-1: Participation in Study by State and Province........................................... 49 Figure 5.4.2-1: Plot of Micro-Deval Control Sample Results .......................................... 61 Figure 5.5.2-1: Plot of Magnesium Sulfate Soundness Control Sample Data.................. 65 Figure 5.7.1-1: Plot of Canadian Freeze-Thaw Cooling Rates......................................... 71 Figure 5.7.1-2: Brownwood Control Samples .................................................................. 72 Figure 5.9.2-1: Comparison of Aggregate Crushing Value Tests Results ...................... 77 Figure 5.12.1-1: Apparatus for Measuring the Shape Ratios for Aggregate Particles.................................................................................................... 82 Figure 6.1.3-1: Climatic Regions of the United States (Desolminihae, Hudson, and Ricci) ................................................................................................. 88 Figure 6.2.1-1: Micro-Deval vs. Performance.................................................................. 89 Figure 6.2.1-2: Magnesium Sulfate Soundness vs. Performance ..................................... 91 Figure 6.2.1-3: L.A. Abrasion vs. Performance................................................................ 92 Figure 6.2.1-4: Canadian Freeze-Thaw Soundness vs. Performance ............................... 94 Figure 6.2.1-5: Aggregate Crushing Value vs. Performance............................................ 95 Figure 6.2.1-6: Aggregate Crushing Value (SSD) vs. Performance................................. 97 Figure 6.2.1-7: Absorption vs. Performance..................................................................... 98 Figure 6.2.1-8: Specific Gravity (Bulk) vs. Performance............................................... 100 Figure 6.2.1-9: Particle Shape Factor vs. Performance .................................................. 101 Figure 6.2.1-10: Percent Crushed (1+) vs. Performance .................................................. 103 Figure 6.2.1-11: Percent Crushed (2+) vs. Performance .................................................. 104 Figure 6.2.2-1: L.A. Abrasion vs. Micro-Deval ............................................................. 106 Figure 6.2.2-2: Magnesium Sulfate Soundness vs. Micro-Deval ................................... 108 Figure 6.2.2-3: Canadian Freeze-Thaw vs. Micro-Deval ............................................... 109 Figure 6.2.2-4: Aggregate Crushing Value vs. Micro-Deval ......................................... 111 Figure 6.2.2-5: Aggregate Crushing Value (SSD) vs. Micro-Deval .............................. 112 Figure 6.2.2-6: Absorption vs. Micro-Deval .................................................................. 114 Figure 6.2.2-7: Specific Gravity (Bulk) vs. Micro-Deval............................................... 115 Figure 6.2.2-8: Particle Shape Factor vs. Micro-Deval .................................................. 117 Figure 6.2.3-1: Aggregate Crushing Value vs. Canadian Freeze-Thaw......................... 119 Figure 6.2.3-2: Particle Shape Factor vs. Canadian Freeze-Thaw.................................. 120 Figure 6.2.5.1-1: Micro-Deval vs. Performance................................................................ 123 Figure 6.2.5.1-2: Magnesium Sulfate Soundness vs. Micro-Deval ................................... 124 Figure 6.2.5.1-3: L.A. Abrasion vs. Micro-Deval ............................................................. 125 Figure 6.2.5.1-4: Canadian Freeze-Thaw vs. Micro-Deval ............................................... 126 xv

Figure 6.2.5.1-5: Aggregate Crushing Value vs. Micro-Deval ......................................... 127 Figure 6.2.5.1-6: Aggregate Crushing Value (SSD) vs. Micro-Deval .............................. 128 Figure 6.2.5.1-7: Absorption vs. Micro-Deval .................................................................. 129 Figure 6.2.5.1-8: Specific Gravity (Bulk) vs. Micro-Deval............................................... 130 Figure 6.2.5.1-9: Particle Shape Factor vs. Micro-Deval .................................................. 131 Figure 6.2.5.2-1: Aggregate Crushing Value vs. Canadian Freeze-Thaw......................... 132 Figure 6.2.6.1-1: Micro-Deval vs. Performance................................................................ 134 Figure 6.2.6.1-2: Magnesium Sulfate Soundness vs. Micro-Deval ................................... 135 Figure 6.2.6.1-3: L.A. Abrasion vs. Micro-Deval ............................................................. 136 Figure 6.2.6.1-4: Canadian Freeze-Thaw vs. Micro-Deval ............................................... 137 Figure 6.2.6.1-5: Aggregate Crushing Value vs. Micro-Deval ......................................... 138 Figure 6.2.6.1-6: Aggregate Crushing Value (SSD) vs. Micro-Deval .............................. 139 Figure 6.2.6.1-7: Absorption vs. Micro-Deval .................................................................. 140 Figure 6.2.6.1-8: Specific Gravity (Bulk) vs. Micro-Deval............................................... 141 Figure 6.2.6.1-9: Particle Shape Factor vs. Micro-Deval .................................................. 142 Figure 6.2.6.2-1: Aggregate Crushing Value vs. Canadian Freeze-Thaw......................... 143 Figure 6.2.7.1-1: Micro-Deval vs. Performance................................................................ 146 Figure 6.2.7.1-2: Magnesium Sulfate Soundness vs. Micro-Deval ................................... 147 Figure 6.2.7.1-3: L.A. Abrasion vs. Micro-Deval ............................................................. 148 Figure 6.2.7.1-4: Canadian Freeze-Thaw vs. Micro-Deval ............................................... 149 Figure 6.2.7.1-5: Aggregate Crushing Value vs. Micro-Deval ......................................... 150 Figure 6.2.7.1-6: Aggregate Crushing Value (SSD) vs. Micro-Deval .............................. 151 Figure 6.2.7.1-7: Absorption vs. Micro-Deval .................................................................. 152 Figure 6.2.7.1-8: Specific Gravity (Bulk) vs. Micro-Deval............................................... 153 Figure 6.2.7.1-9: Particle Shape Factor vs. Micro-Deval .................................................. 154 Figure 6.2.7.2-1: Aggregate Crushing Value vs. Canadian Freeze-Thaw......................... 155 Figure 6.2.8.1-1: Micro-Deval vs. Performance................................................................ 156 Figure 6.2.8.1-3: Micro-Deval vs. Performance................................................................ 158 Figure 6.2.8.2-1: Micro-Deval vs. Performance................................................................ 159 Figure 6.2.8.2-2: L.A. Abrasion vs. Micro-Deval ............................................................. 160 Figure 6.2.8.2-3: Aggregate Crushing Value vs. Micro-Deval ......................................... 161 Figure 6.2.8.2-4: Aggregate Crushing Value (SSD) vs. Micro-Deval .............................. 162 Figure 6.2.8.3-1: Micro-Deval vs. Performance................................................................ 164 Figure 6.2.8.4-1: Micro-Deval vs. Performance................................................................ 165 Figure 6.3.1-1: Micro-Deval vs. Performance................................................................ 167 Figure 6.3.1-2: Magnesium Sulfate Soundness vs. Performance ................................... 168 Figure 6.3.1-3: L.A. Abrasion vs. Performance.............................................................. 169 Figure 6.3.1-4: Canadian Freeze-Thaw vs. Performance ............................................... 171 Figure 6.3.1-5: Aggregate Crushing Value vs. Performance.......................................... 172 Figure 6.3.1-6: Aggregate Crushing Value (SSD) vs. Performance............................... 173 Figure 6.3.1-7: Absorption vs. Performance................................................................... 174 Figure 6.3.1-8: Specific Gravity (Bulk) vs. Performance............................................... 175 Figure 6.3.1-9: Particle Shape Factor vs. Performance .................................................. 176 Figure 6.3.2-1: Magnesium Sulfate Soundness vs. Micro-Deval ................................... 178 Figure 6.3.2-2: L.A. Abrasion vs. Micro-Deval ............................................................. 179 xvi

Figure 6.3.2-3: Canadian Freeze-Thaw vs. Micro-Deval ............................................... 180 Figure 6.3.2-4: Aggregate Crushing Value vs. Micro-Deval ......................................... 181 Figure 6.3.2-5: Aggregate Crushing Value (SSD) vs. Micro-Deval .............................. 182 Figure 6.3.2-6: Absorption vs. Micro-Deval .................................................................. 183 Figure 6.3.2-7: Specific Gravity (Bulk) vs. Micro-Deval............................................... 184 Figure 6.3.2-8: Particle Shape Factor vs. Micro-Deval .................................................. 185 Figure 6.3.2-9: Fractured Particles (2 or more sides) vs. Micro-Deval .......................... 186 Figure 6.3.3-1: Absorption vs. Canadian Freeze-Thaw.................................................. 187 Figure 6.3.5.1-1: Micro-Deval vs. Performance................................................................ 188 Figure 6.3.5.1-2: Magnesium Sulfate Soundness vs. Micro-Deval ................................... 189 Figure 6.3.5.1-3: Canadian Freeze-Thaw vs. Micro-Deval ............................................... 190 Figure 6.3.5.1-4: Absorption vs. Micro-Deval .................................................................. 191 Figure 6.3.5.1-5: Specific Gravity (Bulk) vs. Micro-Deval............................................... 192 Figure 6.3.5.2-1: Absorption vs. Canadian Freeze-Thaw.................................................. 193 Figure 6.3.6.1-1: Micro-Deval vs. Performance................................................................ 195 Figure 6.3.6.1-2: Magnesium Sulfate Soundness vs. Micro-Deval ................................... 196 Figure 6.3.6.1-3: Canadian Freeze-Thaw vs. Micro-Deval ............................................... 197 Figure 6.3.6.1-4: Absorption vs. Micro-Deval .................................................................. 198 Figure 6.3.6.1-5: Specific Gravity (Bulk) vs. Micro-Deval............................................... 199 Figure 6.3.6.2-1: Absorption vs. Canadian Freeze-Thaw.................................................. 200 Figure 6.4.2-1: Magnesium Sulfate Soundness vs. Micro-Deval ................................... 202 Figure 6.4.2-2: L.A. Abrasion vs. Micro-Deval ............................................................. 203 Figure 6.4.2-3: Canadian Freeze-Thaw vs. Micro-Deval ............................................... 205 Figure 6.4.2-4: Aggregate Crushing Value vs. Micro-Deval ......................................... 206 Figure 6.4.2-5: Aggregate Crushing Value (SSD) vs. Micro-Deval .............................. 207 Figure 6.4.2-6: Absorption vs. Micro-Deval .................................................................. 209 Figure 6.4.2-7: Specific Gravity (Bulk) vs. Micro-Deval............................................... 210 Figure 6.4.2-8: Particle Shape Factor vs. Micro-Deval .................................................. 211 Figure 6.5.1-1: Micro-Deval vs. Performance................................................................ 214 Figure 6.5.1-2: Magnesium Sulfate Soundness vs. Performance ................................... 215 Figure 6.5.1-3: L.A. Abrasion vs. Performance.............................................................. 216 Figure 6.5.1-4: Canadian Freeze-Thaw vs. Performance ............................................... 217 Figure 6.5.1-5: Aggregate Crushing Value vs. Performance.......................................... 218 Figure 6.5.1-6: Aggregate Crushing Value (SSD) vs. Performance............................... 219 Figure 6.5.1-7: Absorption vs. Performance................................................................... 220 Figure 6.5.1-8: Specific Gravity (Bulk) vs. Performance............................................... 221 Figure 6.5.1-9: Particle Shape Factor vs. Performance .................................................. 222 Figure 6.5.2-1: Magnesium Sulfate Soundness vs. Micro-Deval ................................... 224 Figure 6.5.2-2: L.A. Abrasion vs. Micro-Deval ............................................................. 225 Figure 6.5.2-3: Canadian Freeze-Thaw vs. Micro-Deval ............................................... 226 Figure 6.5.2-4: Aggregate Crushing Value vs. Micro-Deval ......................................... 227 Figure 6.5.2-5: Aggregate Crushing Value (SSD) vs. Micro-Deval .............................. 228 Figure 6.5.2-6: Absorption vs. Micro-Deval .................................................................. 229 Figure 6.5.2-7: Specific Gravity (Bulk) vs. Micro-Deval............................................... 230 Figure 6.5.2-8: Particle Shape Factor vs. Micro-Deval .................................................. 231 xvii

Fractured Particles (2 or more sides) vs. Micro-Deval .......................... 232 Aggregate Crushing Value vs. L.A. Abrasion ....................................... 233 Aggregate Crushing Value (SSD) vs. Canadian Freeze-Thaw.............. 234 Absorption vs. Aggregate Crushing Value ............................................ 235 L.A. Abrasion vs. Aggregate Crushing Value ....................................... 240 Aggregate Crushing Value (SSD) vs. Aggregate Crushing Value ........ 241 Specific Gravity (Bulk) vs. Absorption ................................................. 242 Specific Gravity (SSD) vs. Specific Gravity (Bulk).............................. 242 Specific Gravity (Apparent) vs. Specific Gravity (Bulk) ...................... 243 Specific Gravity (Apparent) vs. Specific Gravity (SSD)....................... 244 AIMS Angularity Index vs. Performance .............................................. 252 AIMS Texture Index vs. Performance for All Aggregates .................... 253 AIMS Texture Index vs. Performance with One Aggregate Removed ................................................................................................ 253 Figure 6.7.3-4: AIMS Angularity Index vs. AIMS Texture Index Before MicroDeval ...................................................................................................... 254 Figure 6.7.3-5: AIMS Angularity Index vs. AIMS Texture Index Before MicroDeval ...................................................................................................... 255

Figure 6.5.2-9: Figure 6.5.3-1: Figure 6.5.3-2: Figure 6.5.3-3: Figure 6.6.2-1: Figure 6.6.2-2: Figure 6.6.2-3: Figure 6.6.2-4: Figure 6.6.2-5: Figure 6.6.2-6: Figure 6.7.3-1: Figure 6.7.3-2: Figure 6.7.3-3:

xviii

Chapter 1: Introduction
1.1 Need for Project Today there is an incentive to use more marginal aggregates while still providing long-lasting quality roads and structures. This is due to the depletion of aggregate resources, competition for natural researches from urbanization, and resistance to the opening of new quarries (Senior and Rogers, 1991). Therefore, qualification tests that are able to accurately and consistently discriminate between satisfactory and unsatisfactory aggregate are needed. However, researchers have shown that traditional testing methods are not always suitable alone or even in combination with other tests. As a result, transportation-related agencies have begun internal, independent projects to determine the ability of micro-Deval to successfully determine potential aggregate performance. Several agencies have published reports showing micro-Deval to be an outstanding indicator of field performance. Others have found results that show microDeval as having poor or mixed correlations with field performance. Early research has shown that micro-Deval can, in fact, successfully determine aggregate performance, but differing specification limits are being presented. With the interdependence of

government transportation agencies and aggregate producers nationwide, all parties would benefit from universal recommendations for micro-Deval limits for specifications. Therefore, a comprehensive national research project is needed to determine appropriate micro-Deval limits. 1.2 Project Objective The objective of ICAR project 507 is to determine the significance and application of micro-Deval by determining the ability of micro-Deval to predict an 1

aggregates field performance either alone or in combination with other popular aggregate qualification tests. The tests studied in conjunction with micro-Deval are magnesium sulfate soundness, Los Angeles abrasion, Canadian freeze-thaw, aggregate crushing value, absorption, specific gravity, particle shape factor, and percentage of fractured particles. The field applications investigated for ICAR project 507 were hotmix asphalt, Portland cement concrete, base course, and open-graded friction course. In addition to analyzing all applicable aggregate sources for each field application, subsets based on geological rock type are studied for further correlations more specific to the type of aggregate. 1.3 Scope of Project In order to gain national acceptance of this research using the resources available to this project, the investigation of the most commonly used aggregate qualification tests using a national pool of aggregate was required. Contacts were made with the

departments of transportation across the United States and the provinces throughout Canada to determine the use of current aggregate tests and to obtain aggregate samples. Emphasis was placed on obtaining aggregate samples with a wide geological and mineralogical reach. Once these aggregates were obtained they were processed and prepared according to typical industry standards and tested according to the current accepted test specifications. All of this work was conducted at the Pickle Research Campus of The University of Texas at Austin. In addition, each geological rock type was determined by a trained petrographer, and each aggregates field performance was investigated and determined by ICAR 507. An analysis was then conducted by comparing and

manipulating the tests results to determine correlations between the tests and aggregate field performance. 2

Chapter 2: Review of Literature

2.1 Introduction Before beginning this research a thorough review of the literature was conducted. Knowledge of current aggregate tests and applicable testing methods and factors affecting variability of these tests and methods was needed in order to conduct an efficient research project that would be applicable to the needs of those who provided funds for this research. Two main aggregate qualities must be evaluated from aggregate tests: the susceptibility to wear and impact and the susceptibility to weathering, or durability (Hudec, 1983). To evaluate these qualities aggregate tests must be able to: provide correlations with field performance; be suitable for aggregates from different sources; results in reasonable variability of aggregate from a single source; and give results representative of the whole sample (Shergold, 1948). The following is a review of the literature pertinent to this study, and discussions focus on the qualities of aggregates and aggregates tests mentioned above. 2.2 Field Performance Criteria Two recent research efforts outlined performance qualifications for aggregates in order to compare aggregate field performance with laboratory tests. Research at the Ministry of Transportation of Ontario used a method for defining aggregates as good, fair, and poor (Senior and Rogers, 1991). Their definition of good was 10 years of life beyond which any aggregate still performing well in service. Significant disintegration after only one winter separated poor from fair aggregate. Only one instance of poor or fair performance was needed to be classified as such. The classification system is outlined in Table 2.2-1. 3

Table 2.2-1: Classification System Used by Rogers and Senior Evaluation Good Fair Description Used for many years with no reported failures, popouts, or other signs of poor durability Used at least once where popouts or some reduced service life had resulted, but pavement or structure life extended for over 10 years Used once with noticeable disintegration of pavement after one winter, severely restricting pavement life

Poor

Another classification system was used in research conducted at the National Center for Asphalt Technology (Wu et al., 1998). Their system was similar to that of Rogers et al. except that the failure mechanisms were specifically mentioned to be problems during construction, popouts, raveling, and potholes. Another significant

difference is that eight years of good performance was needed for an aggregate to be classified as good. In addition, two years of good service were required for an aggregate to be classified as fair. Their classification system is shown in Table 2.2-2. Table 2.2-2: Performance Evaluation Used by Wu et al. Evaluation Good Description Used for many years with no significant degradation problem during construction and no significant popouts, raveling, or potholes during service life Used at least once where some degradation occurred during construction and/or some popouts, raveling, and potholes developed, but pavement life extended for over 8 years. Used at least once where raveling, popouts, or combinations developed during the first 2 years, severely restricting pavement

Fair

Poor

Another performance system was used by the Utah State Department of Highways (Miles, 1972). Trained geologists ranked each of 18 aggregates using a scale according to expected field performance in riprap. This would not be acceptable for this research as a more standardized and dependable method is required. 2.3 2.3.1 Micro-Deval Background The Deval test was developed in France in the 1870s to evaluate aggregate to be used for roads, and it was initially adopted by ASTM in 1908 (Amirkhanian et al., 1991). However, the test was abandoned years ago by most for all purposes except railroad ballast because it had poor correlations with field performance (Rogers et al., 1995). The micro-Deval test was adapted from the Deval test in the 1960s in France (Hanna et al., 2003) and was first introduced to North America by use in Quebec. Throughout the 1990s the Ministry of Transportation conducted extensive research to refine and characterize the test. Research on micro-Deval began in the United States in the late 1990s and continues today. 2.3.2 Significance Extensive research by the Ministry of Transportation of Ontario (MTO) in Canada has shown that the micro-Deval test is an excellent indicator of field performance. Micro-Deval has been reported to perform much better than the attrition test and was an excellent indicator of field performance for fine aggregate (Rogers et al., 1991). In 1998 researchers stated that loss limits for micro-Deval (Latham and Rogers, 1998) for various asphalt, base, and concrete applications that would be far more effective than using current L.A. Abrasion limits. Other researchers found in 2003 that for qualifying 104 northern Ontario aggregates micro-Deval had a 64% success rate, and a combination of 5

micro-Deval and visual inspection yielded a 94% success rate (Cooper et al., 2003). In a report prepared for the Public Works and General Services of Canada, micro-Deval was reported to be far more effective at separating the good from the bad than the Los Angeles abrasion and impact test (Richard and Scarlett, 1997). This superiority of the micro-Deval was explained to be due to the fact that it is a wet abrasion test; poorer quality rock types tend to slake or have reduced strength when wet, and in the field aggregates are rarely dry (Rogers and Senior, 1994). Work also done by MTO has shown that combinations of other aggregate tests with micro-Deval can predict aggregate performance. A paper in 1991 by Senior and Rogers explained attempts to graph the test results of two different aggregate tests on different axes to inspect the relation to field performance (Senior and Rogers, 1991). An example of one of their graphs for concrete aggregate is shown in Figure 2.3.2-1. The results of their research found that, for granular base aggregates, micro-Deval and petrographic examination gave the best indication of field performance. They also found that for Portland cement concrete micro-Deval and the Canadian unconfined freeze-thaw test were the best indicators of field performance.

Figure 2.3.2-1:

Example of Plot by Senior and Rogers

As a result of the successful use of micro-Deval in Ontario and the need for a better aggregate qualifying test, several other North American transportation agencies have investigated the use of micro-Deval with mixed results. Most have reported good correlations between micro-Deval and field performance. In 1998 Kandhal and Parker (Kandhal and Parker, 1998) of National Center for Asphalt Technology (NCAT) reported that tests with 16 aggregates of varying field performances from across the U.S. showed that the micro-Deval and magnesium sulfate soundness tests were the two best indicators of aggregate performance for hot-mix asphalt. Losses of 18% for both tests appeared to separate good and fair aggregates from poor aggregates. Two reports in 2003 of investigations by different transportation agencies showed that micro-Deval was a good indicator of field performance. The Colorado Department 7

of Transportation in an unpublished internal report (2003) stated that micro-Deval is clearly a better indicator of field performance. Nineteen aggregates from across The

Colorado were collected and rated good, fair, or poor by regional engineers.

department found that at a micro-Deval loss of 18 all poor aggregates were identified, and a micro-Deval loss of 15 would also identify two fair aggregates. In addition, the Oklahoma Department of Transportation (Hobson et al., 2003) reported on an internal research effort to determine the ability of micro-Deval to determine field performance. Eighteen sources of aggregate from across Oklahoma with known field performances in hot-mix asphalt were tested. The aggregates were rated as good, fair, or poor. The performance history rating correlated exactly with micro-Deval loss. Despite the several research efforts which show that the micro-Deval is a good indicator of field performance, two agencies have found otherwise. In 2001 the Oregon Department of Transportation studied the ability of micro-Deval to identify seven aggregates which had known performance histories. Some of the aggregates studied had failed with excessive degradation due to studded tire use. The researchers concluded in that study that micro-Deval was not a better indicator of field performance (Hunt, 2001). In an extensive project conducted at NCAT, Cooley et al. investigated 72 aggregates from eight states across the Southeast. The results showed that although correlations existed for some mineralogical types in some states, no correlations existed between micro-Deval and field performance when considering the results of all the aggregates as a whole (Cooley and James, 2003). 2.3.3 Correlations with Other Tests Several attempts have been made to correlate micro-Deval with other aggregate tests. Most commonly, several attempts have been made to use micro-Deval as a 8

replacement for the sulfate soundness tests, and most have been successful. Researchers

at MTO reported that correlations between the magnesium sulfate soundness test and the micro-Deval test were very good with a correlation coefficient of R = 0.85 for 106 coarse aggregates (Senior and Rogers, 1991). This correlation can be seen in Figure 2.3.3-1. Additional research at MTO showed that micro-Deval correlated very well with magnesium sulfate for fine aggregate with much less variability both in and between labs (Rogers et al., 1991). In another study from Canada researchers reported that microDeval has a very good correlation with the magnesium sulfate soundness test (Richard and Scarlett, 1997). In 1999 NCAT reported that for 16 coarse aggregates of varying performance, correlations between micro-Deval and all abrasion tests were poor, but correlations could be found between micro-Deval and soundness tests including an excellent correlation with magnesium sulfate (R = 0.848) (Wu et al., 1999). In a report from Texas Tech University, researchers showed that an excellent correlation (R = 0.8365) existed when comparing magnesium sulfate with the product of micro-Deval and absorption results (Jayawickrama et al., 2001). Despite these reports showing that micro-Deval does

correlate with the SS test, one report stated otherwise. Research at NCAT found that when examining the test results of 72 aggregates from eight states in the Southeast, no correlations could be found between micro-Deval and sodium sulfate soundness (Cooley and James, 2003).

Figure 2.3.3-1:

Correlation Between Magnesium Sulfate Soundness and Micro-Deval as Found by Senior and Rogers

Micro-Deval has also been compared with several aggregate abrasion tests with little success. Several researchers have compared micro-Deval with the Los Angeles abrasion and impact test, and all have found that no correlation exists. Both NCAT studies (Cooley and James, 2003; Kandhal and Parker, 1998) reported that micro-Deval has no good correlations with any abrasion tests including the Los Angeles test. In additional research, it was found that for 40 different aggregates of a variety of mineralogical types there was no correlation could be found between the micro-Deval and the British aggregate abrasion value (Latham et al., 1998). They also found that there was a significant difference between running the micro-Deval wet and dry. 10

2.3.4

Precision and Variables The precision of the micro-Deval test has universally been reported as being

excellent. The Ontario Ministry of Transportation has shown that the micro-Deval test is relatively insensitive to sample grading and has a low within-laboratory variation. Moreover, its inter-laboratory variation was tested with samples of 58 different sands tested in several different labs by several different technicians. The results had a

coefficient of variation of only 3.2% (Rogers et al., 1991). It has also been shown that the variability of the micro-Deval test is low (Wu et al., 1999). While investigating the Quebec version of the micro-Deval test which specifies a sample size of 500 grams, research has shown that the repeatability of micro-Deval test is be improved by using three times more material (Latham and Rogers, 1998). specify 1500 grams. Research has shown that several factors could play a role in determining microDeval loss and precision. Canadian research showed that soaking for longer than an hour had negligible effects, but varying the soaking time under an hour does affect results (Latham and Rogers, 1998). However, another study reported that soaking had no effect on micro-Deval loss (Latham et al., 1998). When comparing the micro-Deval test results of 40 different aggregates of a variety of mineralogical types, results showed that a very good correlation with a coefficient of R2 = 0.994 was found between soaking the samples and not soaking but testing with water. This can be seen in Figure 2.3.5-1. Three studies of degradation tests have shown that aggregate size, shape, and texture affects loss, and one study on the micro-Deval has shown the effect of the sample size on loss. It has been reported that aggregate shape played an important factor in a wet L.A. Abrasion test as crushed gravel consistently yielded higher loss than uncrushed gravel (Ekse and Morris, 1959), In addition, it has been shown in a study of the 11 Therefore, current methods

Washington Degradation test that aggregate size, type, and weathered state influenced wet degradation loss (Goonewardane, 1977). Two researchers have shown that abrasion loss in the gyratory testing machine is dependent upon kind of aggregate, aggregate gradation, and particle shape (Moavenzdeh and Goetz, 1963). In a study investigating the affect of sample size on micro-Deval loss, MTO has shown that the percent loss of microDeval was independent of sample size until the sample reached 2000 grams (Latham and Rogers, 1998). 2.3.5 Current Use Despite spreading popularity and being recommended by National Cooperative Highway Research Program (NCHRP) as the best aggregate test for abrasion (Hanna et al., 2003), personal communications with U.S. DOTs for this project has shown that the use of micro-Deval in North America is not widespread. Of 35 state transportation agencies responding to a survey distributed for this project, only seven use micro-Deval and only four more are currently considering future use. Of the remaining 24, only ten have tried micro-Deval. In addition, only three Canadian provinces use micro-Deval: Ontario, Nova Scotia, and Quebec (Richard and Scarlett, 1997). Those state DOTs not using micro-Deval cite a variety of different reasons: no advantage over current tests; no current problems with aggregate qualification; and no time or money to investigate micro-Deval and set limits.

12

Figure 2.3.5-1:

Correlation Between Wet and Soaked Micro-Deval Loss as Found by Woodside and Woodward

2.4 2.4.1

Los Angeles Abrasion Background The Deval abrasion test was developed in France in the 1870s and was initially

adopted by ASTM in 1908 (Amirkhanian et al., 1991). Due to a lack of correlation with field performance, the Los Angeles abrasion test was created around 1925 (Hveem and Smith, 1964). The LA test was later adopted by ASTM in 1937 because it was felt that L.A. had better relationship to field performance (Amirkhanian et al., 1991). Despite overwhelming research that shows that LA is a poor indicator of field performance, the test has been used almost universally since. Two reports have shown that L.A. has good correlations with the British impact value test and thus should be considered an impact test (Hudec, 1983; Senior and Rogers, 1991). The correlation as shown by Senior and 13

Rogers can be seen in Figure 2.4.1-1. As a result, the name of the test was later changed to the Los Angeles abrasion and impact test.

Figure 2.4.1-1:

Correlation Between Aggregate Impact Value and Los Angeles Abrasion as Found by Senior and Rogers

2.4.2

Significance The research results concerning the significance of the Los Angeles test are quite

varied. Almost no agreement can be conclusively made concerning the ability of the L.A. test to indicate strength; however, several have shown that the L.A. test correlates very well with impact tests and is not an indicator of abrasion resistance. Several have also shown that in any application, the L.A. test is not a good indicator of field performance. 14

A strong correlation has been reported between the crushing value test and the L.A. Abrasion test (Shergold, 1948). Additional research followed by investigating a correlation between L.A. loss and the strength of concrete for gravel aggregates with a loss of 42 to 57% (Jumper et al., 1956). They found that for gravel samples a 1% reduction in L.A. loss corresponded to a 1% loss in concrete strength. However, an attempt was made to find a correlation between L.A. and the unconfined compressive strength of the rock, and no correlation could be found except at high values of L.A. loss (Shakoor and Brown, 1996). In addition, other researchers pointed out that often times aggregates with high L.A. losses are very suitable for high-performance concrete (Laplante et al., 1991). Researchers have, however, successfully shown that the Los Angeles abrasion test correlates well with impact tests. Hudec reported that L.A. has good correlations with the British impact value test, and therefore it should be considered an impact test (Hudec, 1983). Additional Canadian work found that correlations between L.A. Abrasion and the aggregate impact value test were very good (Rogers, 2004; Senior and Rogers, 1991). Two more researchers attempted to find correlations between the L.A. Abrasion test, the French Deval test, the German impact, British impact, and modified Marshal test (Kohler and Nagel, 1972). He found that although the strains of the aggregates are very different with these test methods, they characterize the same or at least similar properties of the aggregates. Good correlations were found. Two researchers have shown that the L.A. test is not able to predict abrasion resistance. In 1958 Smith reported that the L.A. test had no correlation with concrete abrasion resistance as measured by the Davis steel ball, the dressing wheel abrasion apparatus, and the Ruemelin shotblast apparatus (Smith, 1958). The Ontario Ministry of

15

Transportation reported that the L.A. Abrasion test was not a good indicator of aggregate abrasion resistance as measured by the aggregate abrasion value test (Rogers, 1980). Although early work shows correlations between the L.A. Abrasion test and field performance in pavements, a large amount of more recent work has refuted this claim. One researcher reported that there was a definite relation between the loss in the L.A. test and the service record of materials used in concrete and asphalt (Woolf, 1937). In addition, Richard and Scarlett reported that, of all the Canadian provinces responding to a survey, almost all of them reported that aggregates passing L.A. loss limits performed well in pavement applications (Richard and Scarlett, 1997). However, Richard and Scarlett also cited Rogers laboratory research with 106 coarse aggregate samples that had shown that L.A. has no relation to field performance for aggregates with loss limits under 50. More researchers have shown that L.A. loss does not necessarily show the ability of the aggregate to relate to the compressive strength of concrete (Amirkhanian et al., 1992). A report in 1959 documented the failures of certain basalt coarse aggregates in pavements in the state of Washington due to the creation of plastic fines after they had passed aggregate qualification tests (Ekse and Morris, 1959). They concluded that the L.A. Abrasion test and the Deval test were not able to discriminate between good and poor aggregate. In 1984 researchers found that L.A. Abrasion values did not show any correlation with performance of aggregates used in surface mixtures and asserted that aggregates should not be rejected on the basis of the Los Angeles abrasion test alone (Gandhi and Lytton, 1984). Researchers in Canada found that L.A. was a poor indicator of performance for base material (Rogers et al., 1995). Other researchers in Canada have reported that L.A. cannot predict field performance (Senior and Rogers, 1991). Some

16

feel that the L.A. test is good for determining the mechanical breakdown due to handling, but not much else (Meininger, 2004; Rogers et al., 2003). In 1993 an attempt was made to determine if aggregate absorption could be estimated utilizing the test results of the Los Angeles abrasion test (Esa et al., 1993). Although no correlations could be found, two conclusions were drawn: aggregates with higher L.A. results would likely have a higher tendency for water absorption; and the relationship of L.A. loss to absorption depends on the aggregate type and is site specific. Research in 1986 attempted to find correlations between L.A. loss, unconfined compressive strength, absorption, and dry density (Mirza, 1986). No correlations were found except a minor correlation between L.A. and unconfined compressive strength at low values of L.A. loss. 2.4.3 Precision In research in the state of California, 12 different laboratories conducted tests on aggregate samples to determine the precision of each. Studying aggregates with a L.A. Abrasion loss of 13 to 18, the researchers found that variability was low with a singleoperator variance of 1.1 and a multi-laboratory variance of 3.53. The largest source of error cited for the Los Angeles abrasion test was different laboratory equipment from lab to lab (Benson and Ames, 1975). 2.4.4 Current Use The Los Angeles abrasion test is the most universally used aggregate qualification test throughout the world (Meininger, 1994). Almost all of the respondents to a national survey indicated that the L.A. Abrasion loss should be a specification requirement and that they were satisfied with the value that their agency had adopted (Amirkhanian et al., 1991). Another survey showed that all but three states use the L.A. test for aggregate 17

qualification, and only two states have a degradation requirement besides the L.A. test (Wu et al., 1998). A report in 1998 for COST 337 and the Laboratoire Central des Ponts et Chausses in Nantes, France documented European use of aggregate tests (Hornych, 1998). The Los Angeles abrasion test was the most widely used abrasion test in Europe with 14 countries. Despite this widespread use of and significant previous research concerning the Los Angeles abrasion test, 26% of all US agencies were unaware of where their loss limits originated (Amirkhanian et al., 1992). In addition, confidence in the L.A. results alone was 36%, and confidence when the Los Angeles abrasion test was used in combination with wet abrasion tests or soundness tests was 84%. Some now believe that L.A. principally determines susceptibility to breakdown during handling (Meininger, 2004). 2.5 Other Abrasion Tests Several abrasion tests have been developed over the years to better simulate the degradation that will occur in the field. After the invention of the Los Angeles abrasion test, the majority of the tests developed involved some form of wet abrasion. This is most likely due to the fact that most became aware that aggregates are not dry in the field and behave differently when wet, and that the current dry abrasion methods were insufficient. After failure of coarse aggregates in Virginia that had passed all qualification tests, the Virginia Department of Highways attempted to create a new test that is strikingly similar to the micro-Deval (Melville, 1948). The test consisted of placing the aggregate in a rotating porcelain jar with 3000 grams of flint pebbles and 1000 grams of distilled water. They authors found that aggregate with poor field performance

consistently had losses higher than 65%. 18

The U.S. Army Corps of Engineers reported attempting to create a wet L.A. test (U.S. Army Corps of Engineers, 1957). Water was added to the drum, and the test was run as normal; however, the results were not significantly different than that of the dry L.A. test. For 26 of 38 aggregates, the wet abrasion results were slightly higher, e.g. 1 to 11%. However, for nine aggregates the dry abrasion loss was 9% higher. A very good correlation could be made between the two by using what was almost a 45 degree line. Other researchers have also proposed several changes to the L.A. test by creating a wet L.A. test (Larson et al., 1971). They proposed running the sample dry for 250

revolutions, and then running the sample wet for 250 revolutions. The loss would then be determined by sieving over #16 test sieve instead of the required #12. In 1959 researchers presented the development a wet attrition test that was also strikingly similar to the micro-Deval (Minor, 1959). The test consisted of rotating 1000 grams of aggregate in a 1-gal plastic jar with 200 cm3 of water. The loss was calculated by pouring the wash water in the sand equivalent cylinder and measuring the settlement after 20 minutes. This test came to be known the Washington Degradation Test, and the authors felt that it produced similar degradation to that found in the field, especially when considering the amount of plastic fines created during handling and use. Preliminary work presented by the authors showed that the better quality aggregates seem to have values above 60. Three years later the Washington degradation test was reported to have a good correlation with performance (Larson et al., 1971). Work in California in 1975 attempted to find a replacement for the sodium sulfate soundness test, and several new test methods were created (Spellman et al., 1975). The first method created was a process which bounced aggregate off a metal surface in an attempt to separate harder from softer particles. The test was quickly deemed

unacceptable since hard but brittle aggregate would break when impacting the metal 19

sheet. Heavy media separation, which uses specific gravity to separate aggregate of differing densities (therefore assuming different qualities), was also evaluated. This test was also found to be inadequate except for specific aggregate types. Another test, called the detrition test, was created that involved placing aggregate alone with water in a 5-gal paint bucket and shaking it on a paint shaker. This test was determined to show the most promise with initial results. Finally, after aggregate failures in Washington, an attempt was made to create a new test by running the aggregate for four hours in the L.A. machine without the steel charge (Ekse and Morris, 1959). The researchers believed that this test would be

effective in predicting the tendency of certain aggregate to create plastic fines in service. They did find that operating the test produced larger amounts of plastic fines, but no correlations with field performance were mentioned. 2.6 2.6.1 Soundness Tests Background The sulfate soundness test was created in 1828 by M. Brard attempting to simulate the freezing of water by using the crystallization of sulfate salts (Rogers et al., 1989). In the years since, two forms of the sulfate soundness test have emerged: the sodium sulfate soundness test and the magnesium sulfate soundness test. Beginning in the 1920s, extensive research was conducted to determine the effect of changing variables of the test and the correlation between test results and field performance. Several have debated the ability of the test to predict field performance, but much research has identified and corrected several parameters of the test which lead to great variability.

20

2.6.2

Significance The conclusions of researchers concerning the ability of the sulfate soundness

tests to predict field performance are very mixed. Results spanning decades both support and deny the correlation of sulfate soundness to performance. Some have suggested that poor aggregate have magnesium sulfate soundness losses exceeding 18 to 20% (Kandhal and Parker, 1998; Paul, 1932). Despite this, most do agree that the repeatability of the soundness tests is extremely poor. Several have shown that sulfate tests are good indicators of performance. Researchers determined in 1987 that the four-cycle soundness test was the best among seven laboratory test methods selected for the research (Papaleontiou et al., 1987). Other researchers have reported that magnesium sulfate soundness, combined with microDeval, were the best two indicators of aggregate performance of hot-mix asphalt for 16 aggregates of varying field performances from across the U.S. (Kandhal and Parker, 1998). They found that losses of 18% for both tests appeared to separate good and fair aggregates from poor aggregates. The Utah State Department of Highways found in 1972 that the sodium sulfate soundness test had the highest correlation for 18 aggregates with a field performance rating that consisted of having a trained geologist rate the aggregate on a scale (Miles, 1972). In addition, a researcher has reported a direct relationship between magnesium sulfate and field performance for fine aggregate (Paul, 1932). Aggregates with a 20% loss or higher performed poorly. Moreover, an

investigation of the sodium sulfate soundness test to determine major problem areas indicated that high coarse aggregate soundness loss was, in fact, found to relate to nondurable aggregates (Sheftick, 1989). Despite several researchers reporting good correlations with field performance, several others have reported that sulfate soundness tests cannot predict field performance. 21

Although the sulfate tests were once the only tool for modeling the pressures of freezing and thawing, the fact that the crystal growth of salts in the pores of aggregates does not adequately simulate environmental conditions is now known (Kandhal and Parker, 1998). Research has shown that the sodium sulfate soundness does not relate to the dilation of slow-cooled concrete specimens made with ten gravels, six limestones, and four sandstones chosen from throughout Pennsylvania (Harman et al., 1970). Despite citing four references which show correlations with field performance, one researcher reported that sulfate soundness tests did not necessarily reflect field performance because stringent limits have been placed on a test that does not adequately model the actual field conditions of aggregate (Bloem, 1966). He explained that several sound aggregates have been rejected by sulfate soundness tests, and several unsound aggregates have been accepted by sulfate soundness tests and caused severe degradation in concrete. He concluded that sulfate soundness tests may be used to accept aggregate but should not necessarily be used to reject them. Another researcher has also argued that sulfate soundness tests should be used to accept aggregates but not reject them, but that due to the large number of laboratories using the test elimination is not possible (Woolf, 1937). Others have also found that the sulfate soundness tests have poor correlation with field performance (Rogers et al., 1989; Senior and Rogers, 1991). Several researchers over the years have presented reasoning for use of the magnesium sulfate test as compared to the sodium sulfate test. The variation of solubility of magnesium sulfate in the temperature range of testing is much less than sodium sulfate (Garrity and Kriege, 1935). In addition, sodium sulfate has three crystalline forms at the testing temperature making solution preparation difficult (Larson et al., 1964; McCown, 1932). These issues make the magnesium sulfate test a much more precise and stable test (Meininger, 2004; Wuerpel, 1938), and one researcher reported that in 1938 New York 22

and New England states use magnesium sulfate because it is more stable and required less equipment (Wuerpel, 1938). Some have even called for a complete change of all agencies from sodium sulfate to magnesium sulfate. One researcher called for

magnesium sulfate to be used in place of sodium sulfate due to less variability and a greater severity of attack (Wuerpel, 1939), and Ira suggested a change from sodium sulfate to magnesium sulfate due to the less difficulty in conducting the tests (Paul, 1932). Several studies have been conducted to determine the influence of variations within the test procedure and within the materials used, and these have outlined the difficulty in operating sulfate tests. The effect of temperature on the solubility of

magnesium sulfate and sodium sulfate is well known and discussed in numerous papers and chemistry books. Two researchers reported examining the effects of varying test requirements within the specification limits on the results of both the magnesium sulfate and sodium sulfate tests (Walker and Proudley, 1936). Factors investigated were the effect of sample size, the effect of completeness of drying, the effect of salt grade and brand, and the effect of temperature variation. They found that even minor variations of these factors within the test specification resulted in large variations in the results. Although the specifications were less stringent at that time, the results of the study outlined the precise manner in which the test must be conducted. Similar work at the New York State Department of Transportation showed that minor variations in the solution of the magnesium sulfate test would drastically affect results (Litts, 1981). He suggested carefully monitoring such things as the solution temperature, solution specific gravity, air temperature, and evaporation during drying. One researcher has reported that the use of pans yielded higher loss than using steel wire baskets (Sheftick, 1989), and others concluded that wire baskets should be used over any 23

other container wherever possible (Garrity and Kriege, 1935).

However, personal

communications with a TxDOT representative yielded that steel pans produce no different results (Crenshaw, 2004). The temperature of the aggregates before

reintroduction into the solution has also been shown to greatly affects results (Sheftick, 1989). One researcher has shown that changing the sample gradations within the

specification limits greatly affected sodium sulfate loss (Woolf, 1953), and another reported that drying samples past the time of dehydration affects results (Winslow, 1994). As explained in the section concerning freeze-thaw durability testing, absorption properties play a significant role in durability tests. Therefore, a research project was conducted to determine the relationship of absorption to the sulfate soundness tests (Woolf, 1927). Although no correlation between absorption and sodium sulfate could be found, 82% of aggregates with 2% or higher absorption were unsound according to sodium sulfate soundness, and 100% of aggregates with an absorption value of 4% were unsound according to sodium sulfate soundness. 2.6.3 Precision Several have reported poor repeatability for the magnesium sulfate and sodium sulfate tests for both intra and inter-laboratory testing (Garrity and Kriege, 1935; Larson et al., 1964; Rogers et al., 1991; Walker and Proudley, 1936), but the sodium sulfate test has the highest variability of the two (Benson and Ames, 1975; Meininger, 2002). The cause of this variability results from several aspects of the test: the large effect of minor changes in the temperature the solubility of the solution, the effect of inconsistent drying time, the effect of minor changes in the specific gravity of the solution, the effect of minor changes in the sample gradation, the effect of changing the salt grade, and the effect of the completeness of drying (Forster, 1994; Garrity and Kriege, 1935; Litts, 1981; Sheftick, 1989; Walker and Proudley, 1936; Woolf, 1953). Due to variability 24

issues, the validity of sulfate soundness has been seriously questioned (Meininger, 2002). He added that some aggregates are attacked chemically by the sulfate resulting in erroneously high losses. 2.6.4 Current Use The sulfate soundness test is one of the most widely used aggregate qualification tests in North America. Thirty of 43 states reported using the AASHTO T 104 version of the test (Hanna et al., 2003). In addition, personal communications with DOT

representatives have yielded that a few other variations of the soundness test are also conducted. The sodium sulfate soundness version of the test is the most popular version (Hanna et al., 2003), but personal communications with DOT representatives have shown that a significant number of transportation agencies use the magnesium sulfate version including the Ontario Ministry of Transportation, the Texas Department of Transportation, and the Departments of Transportation of the New England states. 2.7 2.7.1 Freezing and Thawing Background In the early stages of testing freezing and thawing capabilities were not available, and the sulfate soundness tests simulated the freezing and thawing of aggregates as nearly as possible. However, two unconfined freeze-thaw tests have since been developed and have received minor popularity. The Iowa Department of Transportation developed a freeze-thaw test for coarse aggregates which they adopted into their Standard Specifications in 1948 (Marks and Dubberke, 1982). This test exists today as AASHTO T 103 (AASHTO, 2000), which is essentially in the same form as it was in 1948, and has three procedures: 50 cycles of freezing and thawing with the aggregates soaked for 24 hours, 25 cycles with the aggregates vacuum saturated with water, and 16 cycles with the 25

aggregates vacuum saturated with ethyl alcohol. Another unconfined freeze-thaw test has been more recently developed by the Ministry of Transportation using over 20 years of internal research (Rogers et al., 1989). This extensive research has led to several

modifications to the test which has resulted in better correlations with field performance. This test consists of five cycles of freezing and thawing after the aggregates have been soaked for 24 hours in a 5% sodium chloride solution and then drained. Several other tests for simulating freezing and thawing conditions have also been developed but have not received wide acceptance. The Iowa DOT conducted

experiments to predict D-cracking susceptibility by creating the Iowa Pore Index test (Marks and Dubberke, 1982). This test was intended to model the water absorbed in the micro-pores. Water is pressure injected into the aggregate, and the water absorbed in the first 15 minutes after the first minute is recorded. The water absorbed during this time was thought to represent the water absorbed in the micro-pores which would be deleterious. Initial correlations with field performance were good. The Washington hydraulic fracture test consists of pressurizing the aggregate with water and then suddenly releasing the pressure (Issa and Issa, 1999). This results in an internal pressure that, in theory, should represent the internal pressures due to the freezing of water in the pores. However, two researchers reported tests that showed that the Washington hydraulic fracture test would not fracture chert particles which are strongly associated with freeze-thaw durability and especially pop-outs (Embacher and Snyder, 2003). Thus, the test is not adequate. The development of another test for determining the frost susceptibility of aggregates has been reported (Korhonen and Charest, 1995). The test involves cycling aggregate between liquid nitrogen and hot water to simulate freezing conditions. This

26

test appears to rule out frost susceptible aggregates; however, it is quite unique and is not regularly used. 2.7.2 Significance Opinions on the ability of unconfined freeze-thaw tests to predict performance vary. In 1955 Powers discussed some of the basics of freezing and thawing and

explained that some material not broken down while unconfined does break down when frozen in concrete (Powers, 1955). A complicated interaction exists between the Any water being

aggregate and the paste under freezing and thawing conditions.

expelled from aggregate must pass into the surrounding paste, and depending on the paste, failure may or may not occur regardless of the aggregate soundness. In addition, other researchers have claimed that unconfined freeze-thaw tests can not model the freeze-thaw durability of aggregates in concrete because they do not model the condition of the aggregate in the paste (Larson et al., 1964). Two more researchers added to the discussion by writing that they felt that unconfined freeze-thaw testing of aggregate is more time consuming than sulfate soundness and provides no more, and possibly less, information than does sulfate soundness (Boucher and Selig, 1987). Some feel that the current AASHTO freeze-thaw test creates an unrealistically harsh environment by subjecting them to temperature swings that are not realistic and by saturating the aggregates beyond the point which will be seen in the field. Field measurements have shown that concrete rarely cools faster than 5 F/hr (Powers, 1955), and concrete specimens in Ontario have been shown to rarely experience a cooling rate over 2 C/hr (Nokken et al., 2004). Several have shown that the cooling rate and minimum temperature drastically affect freeze-thaw durability (Larson et al., 1964; Pigeon et al., 1985; Rogers et al., 1989). Two other researchers have noted the importance of concrete drying time in the field on aggregate durability, and 27

they presented a new confined aggregate test to allow for drying (Verbeck and Landgren, 1960). Actual correlations between unconfined aggregate freeze-thaw tests and field performance have been mixed. One researcher compiled the results of several research studies in the Midwest and concluded that accelerated freeze-thaw tests are valuable since several deleterious aggregate which did not break down using sulfate soundness did break down with freeze-thaw testing (McCown, 1932). However, the unconfined freeze-thaw test in water has been shown to pass some aggregates that are deleterious (Shakoor et al., 1982). In addition, although initial results after the development of the Iowa unconfined freeze-thaw test seemed promising; a replacement was later developed due to dissatisfaction with correlations with field performance (Marks and Dubberke, 1982). However, the Ministry of Transportation of Ontario has found that the Canadian unconfined test yields good correlations with field performance (Rogers et al., 1989). A good correlation with field performance for 100 Ontario aggregates was found by plotting unconfined freeze-thaw with absorption. More Canadian research found that for Portland cement concrete, the Canadian unconfined freeze-thaw test was one of the best indicators of field performance (Senior and Rogers, 1991) and that the Canadian freeze-thaw test, combined with micro-Deval, can determine aggregate performance with only 5% error (Rogers et al., 2003). The differing results relating to freeze-thaw and field performance is even further complicated by the fact that NCAT has reported good correlations between the AASHTO freeze-thaw, the Canadian freeze-thaw, and the sulfate soundness tests (Wu et al., 1999). The effects of changing the parameters of freeze-thaw tests have been well documented. As mentioned, altering the freezing rate drastically affects freeze-thaw loss (Larson et al., 1964; Pigeon et al., 1985; Rogers et al., 1989). Additional work has shown 28

that the development of an unconfined freeze-thaw test which showed that that the cooling rate, solution strength, and minimum temperature all affected loss (Rogers et al., 1989). They found that the optimum cooling rate was about 7.5 C/hr, the optimum solution strength was 3%, and the optimum minimum temp is 18 C. Additional work found that maximum aggregate expansion during freezing occurred when using approximately a 5% NaCl solution (Litvan, 1976). The relationship of pore characteristics and aggregate size to durability is also well documented. The movement of water out of the aggregate, and hence, the durability of the aggregate, is governed by the pore size, porosity, and the aggregate size (Powers, 1955; Verbeck and Landgren, 1960). Aggregates with finer pores and large absorption that can become critically saturated are at high risk. Aggregates with a large pore volume, but with large pores, are typically sound because it is difficult to critically saturate large pores. Aggregates with fine pores but low absorption may not fail because although the aggregate may become critical saturated, there may not be enough force to be detrimental (Stark, 1976). The most feasible way to reduce susceptibility to Dcracking is to reduce the maximum aggregate particle size which decreases the amount of force that must be contained within. All of these observations are in line with those of Powers on frost durability (Powers, 1955). 2.7.3 Current Use The unconfined freezing and thawing tests have not gained popularity. Of 43 states responding to a survey, AASHTO T 161, which tests concrete beams, was the most popular aggregate freeze-thaw test (Hanna et al., 2003). The Canadian unconfined

freeze-thaw test is used in several provinces in Canada, and only seven U.S. transportation agencies conduct unconfined freezing and thawing tests (Volger and Grove, 1989). Of those seven, six use AASHTO T 103 procedure B (freezing and 29

thawing in water). However, no mention of the thawing procedures used is made by Vogler and Grove. Personal communications with multiple transportation agencies

across the U.S. have yielded that if an unconfined freeze-thaw test is used, there is wide variation from state to state. The use of absorption for the determination of aggregate durability is slightly higher with 13 of 43 states using the test (Hanna et al., 2003). However, others have noted that absorption provides an indication of effective porosity, but it does not provide information regarding pore size which has been shown to be crucial to durability (Swenson and Chaly, 1956). One researcher has argued that current ASTM methods for determining absorption are incorrect (Landgren, 1994). Surface drying by toweling can remove water from pores from certain aggregates, especially lightweight aggregates. Evidence of only isolated use of the Washington hydraulic fracture test and the Iowa pore index test was found in the literature and during personal communication with multiple DOT representatives. 2.8 2.8.1 Petrographic Evaluation Background The value of the petrographic evaluation of aggregates was shown during the early stages of aggregate testing in North America. The basic knowledge that certain rock types were more suitable than others for use in construction was known and applied in Ontario, Canada by 1905 (Rogers, 1990). Twenty-three years later a researcher wrote that a few hundred dollars spent for a geologists advice before opening a quarry may save several hundred thousand dollars loss later on since laboratory tests alone had often not been enough to determine potential aggregate performance (Loughlin, 1928).

30

Another stated that freeze-thaw and soundness tests were useless without consideration given to service records and field performance (McCown, 1932). The first known formal laboratory petrographic evaluations of aggregates began in 1933 when William Y. (Bill) Holland, a petrographer at the Bureau of Reclamation in Denver, Colorado, began examining the properties of aggregates which failed sulfate soundness testing to determine why they were failing. By 1934, Holland was studying all aggregates as they arrived at the laboratory (DePuy, 1990). After finding alkali-silicareactions as the cause of serious cracking in Parker Dam in 1940, which is downstream of the Hoover Dam on the Colorado River, the first formal Petrographic Laboratory was established in 1941 at the Bureau of Reclamation to study the problem (DePuy, 1990). In 1944 petrographers were separated from geologists, and petrographic evaluation of aggregates has since evolved into refined systematic processes which provide useful information concerning field performance. In 1956 a method was reported for determining aggregate quality by petrographic analysis (Swenson and Chaly, 1956). The method consisted of systematically inspecting an aggregate for a series of deleterious qualities which should be evaluated by the engineer. These qualities were separated into two groups: physical and chemical. The physical properties noted which pertain to this research were encrustations, highly weathered surfaces, highly polished surfaces, undesirable shapes, extreme fineness, undesirable pore characteristics, high volume change with wetting and drying, lamination and cleavage, soft and weak particles, and unfavorable thermal expansion. Several

important chemical properties were also noted, but these are not within the scope of this research. Swenson and Chaly asserted that examining coarse aggregate for these

qualities would identify the ability of coarse aggregates to perform in various applications. 31

Attempts were made in Canada in the 1950s to develop a numerical method for the petrographic analysis of coarse aggregate, and by 1953 maximum values had been developed for aggregates of various uses. In a paper in 1990 Rogers explained the development and current use of the petrographic number for aggregate qualification purposes in Ontario (Rogers, 1990). The petrographic number is determined by an experienced petrographer examining about 200 particles of each size fraction of the material in consideration. The percentage weight of each mineralogical type present is then inputted into a standardized spreadsheet representing all aggregate types. The

percent weights are then multiplied by a "factor" that represents each mineralogical types quality to find a weighted value for each mineralogical type present. The "factors" by which the percent weights are multiplied range from 0.9 for excellent to a maximum of 6.0 for poor/deleterious. The weighted values for each mineralogical type are then summed to determine the total value for the aggregate sample which is called the petrographic number. Maximum petrographic numbers can be specified for various uses. In 1994 Mielenz described the basics of petrographic examination of natural, artificial, and recycled aggregate (Mielenz, 1994). Valuable information can be obtained by simple methods such as friability in the fingers, ease of fracturing, nature of the fracture, color, odor on fresh fracture, porosity, and an aggregates reaction to a drop of water. Mielenz continues by giving a detailed list of aggregate properties that should be recorded during a petrographic examination. The non-chemically related aggregate

properties that pertain to this research are mineralogical and lithologic composition, particle shape, surface texture, internal fracturing, coatings, porosity, permeability, absorption, density, hardness, strength, and thermal properties (Mielenz, 1994). Currently, several standardized methods for the petrographic evaluation of coarse

32

aggregate are published by ASTM, AASHTO, the Canadian Standards Association, and other standards institutions around the world. 2.8.2 Significance Two researchers have written that the quality of natural aggregates can be determined by the petrographic analysis because aggregate quality depends on the geological processes which formed them (Rhoades and Mielenz, 1948). As a result, several investigations have documented good relationships between the petrographic examination and field performance of coarse aggregates. One researcher has stated that the best indicator of aggregate performance was a petrographic evaluation since standard test methods used in North America were unable to distinguish fair aggregate from good and poor aggregate (Hudec, 1983). Other researchers have found that for granular base petrographic examination and micro-Deval combined gave the best indication of field performance (Senior and Rogers, 1991) and that for northern Ontario aggregates visual inspection was the best way to determine field performance with an 86% success rate (Fistric et al., 2002). Mielenz has reported that petrographic examination may be a better method to determine the quality of aggregates than laboratory tests: "Detailed petrographic examination is the only procedure that permits comparison and correlation of samples with aggregates previously used or tested. . . By relating the sample to aggregates previously used in construction, aggregate indicated to be unsound by standard tests may be found adequate, or conversely, aggregate indicated to be sound in standard tests might be found unsatisfactory for the intended use" (Mielenz, 1994). In addition, the petrographic number, which is a numerical method of petrographic analysis developed and used in Canada, has been shown to be an excellent tool in determining concrete aggregate performance when combined with other tests, especially freeze-thaw (Rogers, 1990). 33

Although many have shown that petrographic analysis is an excellent tool for qualifying coarse aggregate, some have asserted it should only be used as a supplement to standard laboratory tests. Some have reported that petrographic analysis could

reasonably explain the performance of ballast materials; however, they also stated that petrographic analysis was not yet ready for qualification purposes, but neither were normal index tests (Boucher and Selig, 1987). In a description of the purpose of

petrographic analysis, one researcher wrote that although petrographic analysis should only be used as a supplement for other tests, it is a means for establishing properties and probable performance of concrete aggregate by establishing the fundamental nature of aggregate (Mielenz, 1994). Another has also stated that although petrographic

examinations are one of the best methods for determining aggregate performance when conducted intelligently, they are not sufficient alone, and emphasis should be placed on comparison with other laboratory tests and field performance (Rogers, 1990). 2.8.3 Precision Rogers has reported that when technicians are asked to classify a large number of aggregates, agreement would be reached 85% of the time (Rogers, 1990). This accuracy is accomplished by using trained petrographers and holding periodic seminars to improve classification techniques. No other literature concerning the precision of petrographic investigations could be found; however, due to the subjective nature of the test, error would be expected to be common. 2.8.4 Current Use NCHRP reported that the most commonly used test for the petrographic analysis of concrete aggregates was ASTM C 295 (Hanna et al., 2003). However, only 10 of 41 respondents reported using this test. The petrographic number is used in various 34

Canadian provinces. Personal communications with several transportation agencies have yielded that the methods used for petrographic analysis vary widely if petrographic analysis is used at all. 2.9 2.9.1 Strength and Impact Significance Although no AASHTO or ASTM methods exist for directly determining aggregate strength (Hanna et al., 2003), several other countries have standardized methods for aggregate qualification use that have been used for some time. The British aggregate crushing value (British Standards Institution, 1990) consists of crushing a sample in a compression machine under 400 kN of load and then measuring the deflection and sieving to determine a loss. The British aggregate impact value (British Standards Institution, 1990) is also frequently used in Great Britain for the determination of an aggregates resistance to impact. The aggregate impact value test consists of repeatedly dropping a weight onto a small sample and then sieving to determine the loss. A wet version of the test comes in the form of an Australian standard (Standards Association of Australia, 1980). Different crushing strength values can be obtained by testing the aggregate in both oven dry and saturated surface dry conditions. The wet crushing test has been shown to be useful in evaluating aggregate strength when evaluating both the strength of the aggregate and the fines produced. Also of importance is the variation between the aggregate crushing value results and wet crushing value results for a given aggregate. Larger variations between the aggregate crushing value and wet crushing value have been shown to correlate with aggregate performing poorly due to wetting and drying and freezing and thawing (Goldsworthy, 2004).

35

Several researchers have investigated correlations between the dry aggregate crushing value test, the aggregate impact value test, and the Los Angeles abrasion and impact test. One reported a strong correlation between the aggregate crushing value test and the Los Angeles abrasion test (Shergold, 1948). Canadian research has found that correlations between the L.A. test and the aggregate impact value test were very good (Senior and Rogers, 1991). Another Canadian researcher stated that the L.A. Abrasion test was found to have good correlations with the impact value test and for that reason is now considered an impact test (Hudec, 1983). In a wide reaching investigation in 1972 of impact and abrasion tests used in Europe, one researcher found correlations between the L.A. Abrasion test, the French Deval test, the German impact, British impact, and modified Marshall test (Kohler and Nagel, 1972). Other work in the United Kingdom has examined the method of degradation and significance of the aggregate crushing value and aggregate impact value tests. One researcher reported that the aggregate impact value is influenced primarily by the petrographic characteristics of aggregate, but the particle shape, especially the flakiness, plays a large factor (Ramsay, 1965). Others have shown that there is a direct relationship between aggregate flakiness and aggregate crushing value and aggregate impact value strengths (Dhir et al., 1971). Testing aggregates multiple times showed that once the flakiness of an aggregate was decreased, the performance as measured by these tests improved. In addition, they found a good correlation between the aggregate impact value and the aggregate crushing value. Finally, other British researchers claimed that although the aggregate crushing value test had gained more popularity, results show that a wet version of the aggregate impact value provides a fuller understanding of the strength of the aggregate (Ramsay et al., 1977).

36

2.9.2

Current Use Variations of the British aggregate crushing value have been used for some time

in Great Britain, Australia, and New Zealand. This test is considered valuable and used for qualification purposes in these countries (Goldsworthy, 2004), and the aggregate crushing value was reported by Hanna et al. as being a reasonable approach for determining aggregate strength (Hanna et al., 2003). The wet version of the aggregate crushing value is used for aggregate qualification purposes in Australia and in New Zealand. 2.10 Absorption

2.10.1 Significance The relationship of pore characteristics and aggregate size to durability is well documented. The movement of water out of the aggregate, and hence, the durability of the aggregate, is governed by the pore size, porosity, and the aggregate size (Powers, 1955; Verbeck and Landgren, 1960). Aggregates with finer pores and large absorption that can become critically saturated are at high risk. Aggregates with a large pore volume, but with large pores, are typically sound because it is difficult to critically saturate large pores. Aggregates with fine pores, but low absorption, may not fail

because although the aggregate may become critical saturated, there may not be enough force to be detrimental (Stark, 1976). Several researchers have attempted to determine correlations between absorption and other aggregate qualifying tests. A research project was conducted to determine the relationship of absorption to the sulfate soundness tests (Woolf, 1927). Although no correlation between absorption and the sodium sulfate soundness test could be found, 82% of aggregates with 2% or higher absorption were unsound according to sodium 37

sulfate soundness results, and 100% of aggregates with an absorption value of 4% were unsound according to sodium sulfate soundness results. Other attempts have been made to correlate absorption with aggregate qualification tests. An attempt was made to determine if aggregate absorption could be estimated utilizing the test results of the Los Angeles abrasion test (Esa et al., 1993). Although no correlations could be found, two conclusions were drawn: aggregates with higher Los Angeles abrasion results would likely have a higher tendency for water absorption and the relationship of L.A. loss to absorption depends on the aggregate type and is site specific. More recent research attempted to find correlations between L.A. loss, unconfined compressive strength, absorption, and dry density (Mirza, 1986). No correlations were found except a minor correlation between the Los Angeles abrasion test and unconfined compressive strength at low values of L.A. loss. 2.10.2 Current Use Hanna et al. reported that the AASHTO T 85 (Specific Gravity and Absorption of Coarse Aggregate) test was the most used test among 43 survey respondents with 39 citing use (Hanna et al., 2003). However, some researchers have noted that absorption provides an indication of effective porosity, but it does not provide information regarding pore size which has been shown to be crucial to durability (Swenson and Chaly, 1956). Another has argued that current ASTM methods for determining absorption are incorrect (Landgren, 1994). Surface drying by toweling can remove water from pores from certain aggregates, especially lightweight aggregates.

38

2.11

Aggregate Shape and Surface Texture

2.11.1 Significance Aggregate shape, angularity, and surface texture have been shown to affect the results of degradation tests (Ekse and Morris, 1959; Goonewardane, 1977; Moavenzdeh and Goetz, 1963), the performance of hot-mix asphalt (Ahlrich, 1996), and concrete (Meininger, 2004). In addition, particle shape has been shown to be a factor in impact and crushing tests (Dhir et al., 1971). The ASTM D 4791 test method for flat and elongated particles is cited as providing a good indication of the effect of particle shape on fresh and hardened concrete properties (Hanna et al., 2003). 2.11.2 Current Use The uncompacted void test AASHTO TP 56 is the most commonly used test for aggregate shape, angularity, and surface texture (Hanna et al., 2003). The ASTM D 4791 test method for flat and elongated particles is also used by several agencies, but the majority of states running the test do so using ratios of the aggregate dimensions (Hanna et al., 2003; Rogers, 2004). Investigations into replacing the flat and elongated test with the new AIMS scanning machine are being conducted by transportation agencies.

39

Chapter 3: Aggregate Acquisition and Preparation

3.1 Introduction In order to conduct this research, utmost importance had to be given to the acquisition of coarse aggregate samples from commonly used quarries around the nation for which field performance was known. Aggregates representing use in the four most common applications, hot-mix asphalt, Portland cement concrete, base course, and opengraded friction course, were sought. To provide nationally accepted and applicable results, a wide range of geological rock types, state participation, and geographic representation was needed. To accomplish this task, the contiguous forty-eight state transportation agencies and the transportation agencies of the Canadian provinces bordering the U.S. were contacted, and requests for aggregate were made. 3.2 Initial Survey One of the first steps was to determine the extent of knowledge and use of the micro-Deval test. A survey (Figure A.1 in Appendix A), was sent to the forty-eight contiguous state Departments of Transportation (DOT), most Canadian provincial transportation ministries, and a few aggregate producers located in the United States. The survey aimed to obtain information on current aggregate qualifying tests in use, confidence in the results of these tests, knowledge of the micro-Deval test, interest in the micro-Deval test, and knowledge of aggregate sources which might be linked to failures in hot-mix asphalt, Portland cement concrete, base course, and open-graded friction course. Ninety-five percent of the surveys were completed and returned. This high response rate, along with the information attained on the form, revealed considerable

41

interest in the micro-Deval test. Every responder indicated interest in receiving project updates and findings as well as a copy of the report upon completion. The responses from the survey indicated that many had heard of the test but were unsure as to the role of micro-Deval testing. The responses ranged from not more than hearing the name micro-Deval a few times to some DOTs such as Colorado and Oklahoma performing a miniature study on the test and Texas (TxDOT) including the test in their general qualifying test suite. Many responders mentioned being familiar with NCAT report 98-4 (Wu et al., 1998) or NCAT report 02-09 (Cooley et al., 2002). 3.3 Aggregate Test Suite Determination The suite of tests for the ICAR 507 project was formed from the responses of the initial survey and the literature review. The initial survey sent asked for aggregate tests currently in use by each agency. Those tests commonly reported were selected. These consisted of tests such as the L.A. abrasion, sulfate soundness, absorption, and specific gravity. The remaining tests of the suite were filled with help from the NCHRP report 420C by Hanna et al. (2003). NCHRP Report 4-20C contains a table and flow chart which help the reader determine which aggregate test will yield certain aggregate properties. The combination of these two sources of information allowed ICAR 507 to determine a suite of tests that could sufficiently characterize each aggregate source. The aggregate property sought with the corresponding aggregate tests is as follows:

42

Property Abrasion

Test Method AASHTO TP 58, Resistance of Coarse Aggregate to Degradation by Abrasion in the Micro-Deval Apparatus AASHTO T 96, Resistance to Degradation of Small-Size Coarse Aggregate by Abrasion and Impact in the Los Angeles Machine

Absorption

AASHTO T 85, Specific Gravity and Absorption of Coarse Aggregate

Mineralogy

ASTM C 295, Guide for Petrographic Examination of Aggregates for Concrete

Shape and angularity

ASTM D 4791, Flat Particles, Elongated Particles, or Flat and Elongated Particles in Coarse Aggregate ASTM D 5821, Determining the Percentage of Fractured Particles in Coarse Aggregate

Specific Gravity

AASHTO T 85, Specific Gravity and Absorption of Coarse Aggregate

Strength

AASHTO T 96, Resistance to Degradation of Small-Size Coarse Aggregate by Abrasion and Impact in the Los Angeles Machine BS 812: Part 110 (British Standard), Aggregate Crushing Value BS 812: Part 110 (variation), Aggregate Crushing Value (SSD) 43

3.4

Field Performance Rating Determination Since the one of the goals of this research is to correlate micro-Deval test results

with the field performance of aggregates, the field performance rating is crucial. Originally, the performance ratings given in the responses to the initial survey were intended to be used as the performance rating of each aggregate source. As the project developed and more published literature was reviewed, it was felt that strictly using the ratings provided in the survey responses could result in a subjective rating system. To help define a more objective rating system, three rating systems were found from past research. These were found in published work by Senior and Rogers (1991), Wu et al. (1998), and Phillips et al. (2000). Senior and Rogers (1991) developed a rating scale for the field performance evaluation criteria of coarse aggregates used in granular base and asphaltic and Portland cement concrete: Good used for many years with no reported failures, popouts, or other

signs of poor durability, Fair used at least once where popouts or some reduced service life had

resulted, but pavement or structure life extended for over 10 years, and Poor used once with noticeable disintegration of pavement after one

winter, severely restricting pavement life. The second published rating scale found was the rating system developed by Wu et al. (1998) which looked into characterizing aggregates used in asphaltic concrete only. Phillips et al. (2000) utilized this system for rating aggregate performance in their study as well. The scale is as follows: Good used for many years with no significant degradation problem

during construction and no significant popouts, raveling, or potholing during service life, 44

Fair used at least once where some degradation occurred during

construction and/or some popouts, raveling, and potholing developed, but pavement life extended for over 8 years, and Poor used at least once where raveling, popouts, or combinations

developed during the first two years, severely restricting pavement use. The third rating system that was found through a literature review was a system by Phillips et al. (2000) which analyzed and classified bituminous pavements. Phillips et al. referenced the work performed by Wu et al. (1998) in their study. The rating system consisted of the following criteria: Good used for many years with no significant aggregate degradation

problem during construction and no significant popouts, raveling or potholes, Fair used at least once where some degradation occurred during

construction and some popouts, raveling or potholes developed, but pavement life lasted for at least 80% of its expected life, and Poor used at least once where raveling, popouts or combinations thereof

developed during the first two years, reducing the expected life of the pavement design life significantly. ICAR 507 utilized the same three-step rating system (good, fair, and poor) that is shared by the studies listed above. The good and fair ratings used by the ICAR 507 project were similar to those used by Senior and Rogers since this project dealt with bases, hot-mix asphalts, Portland cement concretes, and open-graded friction courses. The poor rating was similar to the systems used by Wu et al. and Phillips et al. in that two years separated poor from fair instead of the one year used by Senior and Rogers. The performance rating system that was decided on and used can be seen in Table 3.4-1. 45

Table 3.4-1: ICAR 507 Evaluation Criteria for Determining the Performance Rating of Aggregates Rating Good Fair Description used for 10 or more years with no reported non-chemical problems used at least once where minor non-chemically related failures require minor repairs, but average life extends beyond 10 years Poor used at least once where severe degradation or failure occurred within 2 years of service or during construction which severely inhibits and/or prevents the use of the application

In addition to development of the performance rating scale, a final survey was constructed. This survey consisted of a series of questions about each source including which applications the source was used in and any problems experienced due to its use. The survey can be found in Appendix A. This final survey was conducted via phone calls and directed to the participating DOTs and ministries of transportations. The goal of the final survey was to make the rating system as objective as possible. Following a phone interview, ICAR 507 personnel determined the performance rating based on the information provided. By ICAR 507 personnel determining the field performance rating, all samples could be compared on equal terms. In no case was a sample rated poor that was initially rated good. In some cases though, a sample 46

determined to be in an adjacent category from its provided performance rating, e.g. initially rated good but was determined to be fair. 3.5 Aggregate Acquisition The initial survey indicated which aggregates could be provided. Along with the identification of sources, field performance ratings were also requested for each source. This provided the ICAR 507 project with potential aggregates for testing and indicated gaps would need to be filled by future aggregate requests. It was quickly found that providers were more likely to offer aggregates of good and fair field performance. Naturally, most providers did not want to claim their aggregate source as poor so more effort was put into obtaining aggregates of poor performance. In order to obtain many poor performing aggregates, an understanding was reached that the provided poor source would not be named in the report for fear of potentially damaging the reputation and business of the quarry. All sources would be given an identification tag for the project. The providers would then be given a list of their provided aggregates and the corresponding project identifications for reference and comparison of results. One potential problem faced was that the performance label given to an aggregate source was based on aggregate test results instead of field performance. In talking with many providers, it was found an aggregate source was rated poor because it had high L.A. Abrasion or sulfate soundness test results, for example, during qualification, not because of problems in use. Care was taken to ensure that the performance ratings were based on field performance and not on aggregate tests with poor correlation to field performance. This was accomplished by communicating with the provider and finding out more information about the source. Sources with a performance rating for a particular use based on qualification tests were not included for use in developing correlations. 47

Attempts were also made to ensure there were enough sources for each usage category: hot-mix asphalt, Portland cement concrete, base course, and open-graded friction course. Each use entails different loadings from transportation, mixing, placement, compaction, and traffic loads and different acceptance criteria may be valid. Collection of aggregates of many usages was easily achieved as many sources were used in two or more categories. Table 3.5-1 shows how the aggregates used in this study are broken down by use. Table 3.5-1: Aggregate Qualifications per Application Application Hot-mix Asphalt Portland Cement Concrete Base Course Open-Graded Friction Course Good 52 39 59 17 Fair 26 4 2 2 Poor 20 6 4 3

Finally, samples of numerous mineralogical backgrounds were sought. Based on the work performed by Cooley et al. (2002), it was found that granites, for example, may not need to be subjected to the same acceptance criteria as sandstones. Therefore, samples were accumulated from across the United State and Canada for verification. Figure 3.5-1 shows the geographical range of participation by states and provinces. The division of the rocks based on geological rock type can be seen in Table 3.5-2. The amount of aggregate from each source initially requested was two or three fifty-five gallon drums from each source. This large amount was requested since it was not known initially how many tests would be performed. It was more beneficial to use the performance rating found through actual field performance because of redundancy, cost 48

requirements, and time requirements. Henceforth, aggregates would only be requested if there was an established field performance for each use of the source. Sieving the aggregates showed that only one fifty-five gallon drum of aggregate was needed if the grading could be limited to the aggregate passing the 25-mm (1-inch) sieve and retained on the 4.75-mm (No. 4) sieve with approximately 50% retained on the 12.5-mm (1/2inch) sieve. When this gradation was unable to be provided, calculations were performed to determine the amount of aggregate needed.

Figure 3.5-1: Participation in Study by State and Province

49

Table 3.5-2: Division of Aggregates by Geological Rock Type Number of aggregate sources 71 22 17 9 7 6 6 4 22 14 4 4 16 5 3 2 2 4 2

Mineralogy Sedimentary Limestone Siliceous Gravel Other Gravel Dolomite/Limestone Dolomite Sandstone Other Igneous Granite Basalt Other Metamorphic Gneiss Quartzite Phyllite Serpentinite Other Other

50

Chapter 4: Field Performance Determination

4.1 Introduction As each state has its own method for determining acceptable loss limits for aggregates, each state also has its own method for determining acceptable field performance. Several states must have less stringent field performance expectations due to harsher weather experienced in that region of the country. Moreover, individuals within a states transportation agency sometimes differ on the matter of performance. Since the determination of field performance contains the most bias and variation from state to state, it is clearly the most critical part of this research. Methods were needed for removing the bias and variability from state-to-state, in order to accurately and consistently apply the field performance ratings. 4.2 Rating System Two rating systems used in previous similar research were examined and considered. As discussed in the literature review, Senior and Rogers (1991) and Wu et al. (1998) formulated such performance rating systems for the evaluation of coarse aggregate. Both were simple and effective and provided criteria for systematically and objectively determining the field performance of aggregate by evaluating the time until failure while in service. Therefore, a similar three-tier rating system was developed for use in this project and is shown in Table 4.2-1. The non-chemically related failures mentioned in the table include, but are not limited to, such failures as pop-outs, excessive degradation, and production of plastic fines, rutting, D-cracking, and any other failure relating only to the physical properties of the aggregate.

51

This three-tier system shown in Table 4.2-1 was felt to be able to provide a way for transportation agencies to easily identify applicable aggregate while providing the research team a simple but effective way to objectively rate aggregates. To further remove bias and variability from the aggregate performance rating system, an attempt Table 4.2-1: Performance Criteria Developed for Use in This Project Evaluation Good Fair Description Used for 10 or more years with no reported nonchemically related failures Used at least once where minor non-chemically related failures require repair, but life extends beyond 10 years. Used at least once with severe degradation of failure occurring within 2 years of service or during construction which severely inhibits or prevents the use of the application

Poor

as made to completely remove the transportation agencies from the final rating decisions. A standardized questionnaire was created to be used in discussion with government transportation agency representatives. This questionnaire is included in the Appendix. Project personnel used this questionnaire to obtain information concerning each aggregate source. The following information was obtained through the questionnaire: the applications in which the aggregate was used years in service for each application average daily traffic for each application yearly exposure to freeze-thaw characteristics of failures if applicable the time until failure if applicable

The information obtained by the use of this survey was then applied internally to the field performance criteria in Table 4.2-1 for the final decision concerning the field 52

performance rating. Only information provided by government transportation agencies concerning field performance was considered for this task. Since DOTs are the owners, their performance records were considered to be more inclusive.

53

Chapter 5: Testing
5.1 Introduction Standard methods of testing produced by several institutions exist today to ensure comparable and consistent results. Research has been conducted to determine the most common use and most applicable test methods for various applications (Hanna et al., 2003). The test methods used in this research were those that both directly apply to the physical property being investigated and were commonly used in North America. This chapter discusses aggregate sample preparation and briefly the procedures involved and results of each test. 5.2 5.2.1 Aggregate Sample Preparation Processing With such a large amount of material arriving at the laboratory, an efficient means of preparing the supplied aggregate was needed. This preparation entailed sieving, washing, drying, and bagging and required a large amount of person-hours. Sieving the aggregates was the first step. As each source of aggregate arrived at the lab, the sample was passed over a set of sieves consisting of a 19.0 mm (3/4 inch), 16.0 mm (5/8 inch), 12.5 mm (1/2 inch), 9.5 mm (3/8 inch), 6.3 mm (1/4 inch), and a 4.75 mm (No. 4). These fractions were kept separate in 5-gallon buckets. Each bucket was then emptied onto a wire screen mesh and rinsed with water. The aggregate was agitated to ensure that the entire surface was exposed to the stream of rinse water. Care was taken during this step to ensure that future testing of the aggregates would be composed of the aggregate only, without any harmful deleterious particles, such as clay, that might be adhered to the aggregate. 55

Following a visual inspection to make sure the aggregate particles were clean, the aggregate was placed in a convection type oven set to 110C (230F). The samples were left in the oven until constant mass was achieved at which point they were removed, allowed to cool, and replaced in buckets. The buckets were sealed to prevent absorption of moisture and stored until ready for bagging. This ensured the aggregates tested would be in an oven-dry state. Bagging the aggregate samples was found to streamline testing. Bagging consisted of weighing out the aggregate into the recommended masses for each test procedure, carefully labeling the bag with its contents and applicable test, and storing these samples until ready for testing. Three samples of each provided source were bagged for each test. Three samples for testing were chosen because it would be easier to denote an outlier in the results of each test. The bagging process proved to be a very efficient means of preparing a large amount of samples for testing. Since most of the testing was completed at locations other than the main laboratory, samples were easily selected, transported, and tested. 5.2.2 Standardizing Gradations One of the problems ICAR 507 faced was dealing with such a wide range of gradations. States specify different gradations for different uses. This created a potential problem in comparing the test results of one sample to the test results of another since Selig and Boucher (1990) and Rogers et al. (2003) have shown that the gradation can affect test results. ICAR 507 decided to create uniform gradations that each sample should meet for comparison purposes. This decision was supported by work performed by Vogler and Grove (1989) who found in responses to a survey that nine of sixteen responding states standardize gradations prior to testing. These uniform gradations were mostly specified 56

by the test specifications. In some instances, the grading requirement of the test specification was left open for determination by the user. For instance, the absorption test was completed using only material passing the 12.5-mm (1/2-inch) sieve and retained on the 9.5-mm (3/8-inch) sieve. This size fraction was selected for two reasons: it is a median range for a typical coarse aggregate used in road construction that should provide a representative test result and the micro-Deval sample most commonly used, grading A, is comprised of 50% of material passing the 12.5-mm (1/2-inch) sieve and retained on the 9.5-mm (3/8-inch) sieve. Since this project is investigating other tests in combination with micro-Deval, it was deemed suitable to use this particle size for comparison purposes with micro-Deval test results. 5.2.3 Performing Test Procedures Once the samples were bagged they were ready for testing. Testing was performed as close to the applicable specifications as possible. Every effort was made to ensure that results would be consistent from batch to batch. Attempts were made to have the same person perform testing on every sample for a given test. This helped reduce or eliminate the user variability that might exist for a particular test. Another means acting as a quality control was the use of a control aggregate. As recommended in the micro-Deval test specification (AASHTO TP 58), a sample of the Brechin No. 2 limestone was obtained from the Ontario Ministry of Transportation. This sample was used to ensure the micro-Deval apparatus was yielding accurate results. A local aggregate was also obtained for control testing. A local aggregate was needed because of the limited amount of Brechin No. 2 available from MTO and the high cost of shipping the aggregate across the U.S.-Canada border. Due to similar mineralogy and consistent production, Phillips et al. (2000) recommended using the Vulcan Materials 57

Company Brownwood quarry limestone for control testing. Vulcan kindly donated four, fifty-five gallon drums of aggregate for use as a control aggregate. As mentioned in the previous section, three samples were bagged and tested so if two of the sample results matched up but the third did not, the average of the similar test results would be taken as the result. Since the objective was not to determine the variability of each test, erroneous test results were discarded. In some cases, only two tests were performed on each source. The Aggregate Crushing Value test (BS 812: Part 110), for example, calls for two samples to be tested and if the results differ by more than 7% of the mean, two more samples should be tested. This precision statement was adopted for other tests where a precision statement could not be found. In some instances, not enough material was available for the sample to coincide exactly as the test specified. Some supplied sources were more heavily weighted toward the larger particle sizes and vice versa. Where this was the case and recommended size fractions for samples were unable to be met, the size fractions missing were replaced with the next larger or smaller size fraction. For instance, the L.A. Abrasion test specification (AASHTO T 96), grading B, calls for 2500 g of material passing the 19.0-mm (3/4-inch) sieve and retained on the 12.5-mm (1/2-inch) sieve. Since aggregate for this project was sieved on the 16.0-mm (5/8-inch) sieve, as required by the micro-Deval specification (AASHTO TP 58), the 2500 g size fractions for the L.A. Abrasion test specification was subdivided into 1250 g of material passing the 19.0-mm (3/4-inch) sieve and retained on the 16.0-mm (5/8-inch) sieve and 1250 g of material passing the 16.0-mm (5/8-inch) sieve and retained on the 12.5-mm (1/2-inch) sieve. If one sample did not have enough material retained on the 16.0-mm (5/8-inch) sieve to fulfill the 1250 g, 2500 g of material was used from the material retained on the 12.5-mm (1/2-inch) sieve.

58

5.3 5.3.1

Petrographic Analysis Test Procedures The ICAR 507 project utilized the help of an experienced petrographer for this

analysis (ASTM C 295). A professional geologist agreed to aid the project in the mineralogical determination of the aggregates acquired. At the request of the petrographer, a representative sample of 9.5-mm (3/8-inch) maximum nominal size aggregate was placed in a sealed plastic bag and sent for examination. Larger and smaller size aggregate was included for analysis when deemed necessary. The sample was returned with the results consisting of the mineralogy of the aggregates. 5.3.2 Results The results requested were for the general classification of the aggregate. Since the aggregates were to be grouped into generalized rock types e.g. limestone, dolomite, siliceous gravel, the percentages of the sources constituents were not sought. Where one or more of the constituents may yield a pronounced effect on the micro-Deval loss, that constituent was listed. For example, some gravels contained a sufficient amount of chert, that when exposed to water, become expansive and could possibly create an artificially high micro-Deval loss. 5.4 5.4.1 Micro-Deval Test Procedures The micro-Deval testing was conducted according to the AASHTO TP 58 (Resistance of Coarse Aggregate to Degradation by Abrasion in the Micro-Deval Apparatus) specification (AASHTO, 1999). This specification consists of soaking a previously weighed 1500 5 gram aggregate sample in two liters of water for one hour, and then placing the aggregate and the water in a small stainless steel drum with 5000 59

grams of small steel balls with diameter of 9.5 0.5 mm. The drum is then sealed and rotated at 100 5 revolutions per minute for a specified amount of time depending on the gradation used. Afterwards, the aggregate sample is removed and wet sieved over the No. 4 and No. 16 sieves. The total amount of aggregate retained on both sieves is collected and oven dried to a constant mass at 230 F, or usually overnight. The loss is determined by calculating the percent loss as retained on the No. 16 sieve. A control sample is to be run every 10 tests in both drums to ensure consistent results. Three aggregate gradations are available for testing coarse aggregate in this test, but each requires a different amount of testing time. This is shown in Table 5.4.1-1. Table 5.4.1-1: Gradations for Use in Micro-Deval Testing Gradation A 3/4" 5/8" 5/8" 1/2" 1/2" 3/8" Testing Time: 375 grams 375 grams 750 grams 120 min. Gradation B 1/2" 3/8" 3/8" 1/4" 1/4" No. 4 Testing Time: 750 grams 375 grams 375 grams 105 min. Testing Time: 95 min. Gradation C 3/8" 1/4" 1/4" No. 4 750 grams 750 grams

For the purposes of this research, a micro-Deval machine was purchased from Gilson. Personal communication with representatives at the Texas Department of

Transportation indicated that wearing of the rollers and steel balls occurs and can affect results (Crenshaw, 2004). A special version of the micro-Deval was produced for

TxDOT that includes improvements to mitigate this wearing, and this version was obtained with an educational discount. Despite this wearing issue, the precision of the micro-Deval machine has otherwise been shown to be very good by other researchers (Jayawickrama et al., 2001). Personal communications with Chris Rogers of the Ontario Ministry of Transportation stated that the reason required testing time changes with the gradation is to provide the same results regardless of gradation (Rogers, 2004). A brief internal 60

experiment showed that gradation A and gradation B of the same source would yield micro-Deval loss values that were within 1% of each other. Primarily gradation A was used unless restricted by lack of material in certain sizes. In those cases gradation B was used. For the micro-Deval test, three samples of each source were tested and the mean was accepted unless an obvious outlier existed. Control samples from the Brownwood quarry in Texas were calibrated with Brechin II samples obtained from MTO and were tested in both cylinders every ten tests. The control sample results were monitored to ensure that they remained within the acceptable limits as specified in AASHTO TP 58. 5.4.2 Results The data for the micro-Deval testing can be seen in the Appendix B. A plot of the control sample data can be seen in Figure 5.4.2-1. This plot shows the micro-Deval apparatus yielded consistent results over the period of testing.
Micro-Deval Control Aggregate Vulcan Brownwood Quarry, Brownwood, TX
14.5

14.0

13.5 Brownwood Upper Limit = 13.2% 13.0

Percent Loss

12.5 Brownwood Mean - Grading A = 12.0 Brownwood Lower Limit = 11.7%

11.5

11.0

10.5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

Test Number

Figure 5.4.2-1:

Plot of Micro-Deval Control Sample Results 61

5.5 5.5.1

Magnesium Sulfate Soundness Test Procedures The magnesium sulfate soundness test was conducted according to the AASHTO

T 104-00 specification (Soundness of Aggregate by Use of Sodium Sulfate or Magnesium Sulfate) (AASHTO, 1999). Several sample size increments can be tested depending on the source being evaluated. The sample sizes that correspond to asphalt, concrete, and base material were used for this research. Each source was hand-sieved to refusal according to the specification and prepared according to the gradation shown in Table 5.5.1-1. Table 5.5.1-1: Sulfate Soundness Sample Gradations Sieve Size 3/4" to 3/8" Consisting of: 3/4" to 1/2" 1/2" to 3/8" 3/8" to No. 4 Mass, grams 1000 10 670 10 330 5 300 5

The specification calls for supersaturated solution of magnesium sulfate to be prepared and maintained at a temperature of 70 2 F and a specific gravity between 1.297 and 1.306. The samples were then soaked in this solution for 18 hours, usually over night, and then dried to a constant mass in a forced air oven at a temperature of 230 F, usually during working hours. After drying, the samples were cooled to room

temperature, and re-soaked. Five cycles of soaking and drying were completed. After the 5th cycle, the samples were then washed with water of 110 10 F by introducing hot water at the bottom of a tank and allowing the water to flow over the top of the tank. The wash water was checked with a 0.2 molar solution of barium chloride to ensure that rinsing is complete. 62

After being oven dried, the samples were then re-sieved to determine the loss. The 3/4 in to 3/8 in material was sieved over a 5/16 in sieve, and the 3/8 in to No. 4 material was sieved over a No. 5 sieve. The percent loss of each size was determined, and the final percent loss was determined by using weight factors of 2/3 for the 3/4 in to 3/8 in material and 1/3 for the 3/8 in to No. 4 material. For the purposes of this research reagent grade magnesium sulfate heptahydrate (MgSO47H2O) was purchased from Spectrum Chemical and used for the preparation of the solution. The temperature was accurately maintained by holding the room

temperature well below 70 F (usually around 60 F) and warming the solution with fish tank heaters. Uniform temperature of the solution was ensured by circulating the solution with a pump. Clear plastic cooking trays were used for holding the samples. Holes with a diameter of 1/8 in were drilled in the bottom to permit solution passage. The rinse tank was constructed by modifying a typical drink cooler. The water intake consisting of a water hose and valve inserted into the cooler drain and the water outtake consisting of a water hose and spigot drilled through the container. The water temperature was

maintained using the faucet controls and two Humboldt H-2986 ASTM specification heaters. This magnesium sulfate test setup was constructed with the advice of a

representative of the Texas Department of Transportation (Crenshaw, 2004). The variability of the sulfate tests in general as shown by previous work (Garrity and Kriege, 1935; Rogers et al., 1991; Walker and Proudley, 1936) is very large. In this research, at least three samples of each source were tested by magnesium sulfate with no more than two of the three being tested in the same run. Control samples were included with each run. If the control samples yielded values outside the acceptable limits

provided in the AASHTO test specification (AASHTO, 1999), then a careful examination was given and additional samples were tested if necessary. 63

5.5.2

Results The individual data sheets for the magnesium sulfate soundness tests are included

in the Appendix C. A breakdown of the loss values per sample and the averages of the losses for each source is also shown in the Appendix C. In general, the loss values of the magnesium sulfate soundness test were not as precise as those of the micro-Deval test. However, they were still within the variability requirements of the AASHTO specification. The variance was typically high only for higher loss values. Typically all values obtained for a source were accepted and used when determining the mean. Loss values were discarded only in rare cases where an obvious outlier existed among one of the three values for a source. This occurred when one of the three values was more than twice the other two as was the case with sources 23, 28, and 43. In all cases including these three, a visual inspection of results showed that the final averages obtained were consistent and realistic. Brechin II control samples were tested in each batch. Figure 5.5.2-1 shows a plot of the control samples. The results of the control samples were on the higher end of the allowable range. This was probably due to the method with which the loss was sieved. A gentle machine sieve was chosen to reduce variability between operators. The method of sieving of this machine was a swirling motion with a gentle tap to dislodge the aggregates. No harsh agitation was introduced which would be expected to break up aggregate, but the higher loss values were probably due to the machine sieving. Any type of machine sieve will introduce more agitation that would hand sieving which is the method stated in the specification. Although the values of the control samples were higher than typically expected, their variability among themselves was within acceptable

64

limits. An example of the variability in the magnesium sulfate test can be seen in the difference between the values of the two control samples tested in test number 3.

Magnesium Sulfate Brechin Control Results

20

18

16

Percent Loss

14

12

10

6 0 1 2 3 4 5 Test Number 6 7 8 9 10

Figure 5.5.2-1:

Plot of Magnesium Sulfate Soundness Control Sample Data

Although using a machine sieve varied from the AASHTO specification, the results are not to be negated. Each transportation agency sieves differently when

determining loss: some use very gentle hand sieving; some use violent machine agitation; and some use wet sieving. It would be impossible to reproduce everyones method. An interview with the manager of a prominent U.S. aggregate testing laboratory indicated that the results of this research are probably comparable provided all samples were sieved for loss the same way. Since every sulfate loss in this research was sieved using an 65

identical method, the results of this study are comparable with those of any agency because each can convert and interpret the results (Sheftick, 2005). 5.5.3 Lab Data Comparison Laboratory testing revealed wide variations within multiple runs performed for each source even though exact same conditions were followed in each case. To further the issue, a small side study was performed to determine how test results vary between different laboratories. Therefore, three sources were sent to three other accredited

laboratories to perform magnesium sulfate testing Texas Department of Transportation, Bowser-Morner, and Ontario Ministry of Transportation. Each laboratory uses the same ASTM specification used in ICAR 507 testing. However, final sieve methods and

magnesium sulfate grades vary for each laboratory. Table 5.5.3-1 summarizes the results . Table 5.5.3-1: MSS Test Data and Procedures Summary TxDOT ID# 11 ID# 72 ID# 96 Salt Grade Solution Temp Sieve Time 13.0 9.0 18.0 Reagent 70 F BM 14.3 9.1 15.7 Technical 70 F MTO 8.8 6.0 15.9 Technical 70 F 1 min rocker ICAR 507 15.6 12.1 18.7 USP 70 F 10 min machine

10 min machine Hand

Of particular importance are the final sieve methods since higher losses are expected for machine-sieved procedures. MTOs results are an exception since the employed rocker simulates hand sieving very closely. As Table 5.5.3-1 illustrates, the data supports that conclusion since TxDOT and ICAR 507 results represent the highest losses observed. More importantly, however, the results show that magnesium sulfate 66

test results vary significantly between various laboratories even though the same specifications are followed. 5.6 Los Angeles Abrasion The L.A. Abrasion test was selected due to its popularity among transportation officials as found through the initial survey. Approximately 96% of the responders said they use or have a specification for the L.A. Abrasion test; however, according to a survey by Amirkhanian (1992), 26% of surveyed agencies were unaware where their L.A. Abrasion specification loss limits originated. Studies by Minor (1959), Rogers et al. (1995), and Richard and Scarlett (1997) have shown poor correlations between the L.A. Abrasion test results and field performance. While the L.A. Abrasion test can predict the mechanical breakdown of aggregate in stockpiling, transportation, and construction, it does not correlate well with field performance. The name of the test implies both abrasion and impact; the L.A. Abrasion test correlates well with other impact tests such as the Aggregate Impact value (BS 812: Part 110) and Aggregate Crushing Value (BS 812: Part 110) as shown by Shergold (1948), Hudec (1983), and Al-Harthi (2001). 5.6.1 Test Procedures L.A. Abrasion tests were performed on the Texas Department of Transportation (TxDOT) machine. Approximately fifteen to twenty samples were completed each visit to the TxDOT laboratory. The AASHTO T 96 specification was followed, with most aggregates tested as the Grading B sample. The largest nominal particle size for some aggregates did not meet Grading B requirements, and the sample was therefore prepared as a Grading C. These samples were noted on the data sheets. The same number of revolutions was performed 67

on both gradings with the number of steel charges being reduced for Grading C to account for the smaller particle size in accordance with test specifications. The L.A. Abrasion test calls for a 5000-g aggregate test sample of to be placed in a revolving drum along with a set number of steel charges averaging 46.8 mm (1-27/32 inch) in diameter. The drum picks up and drops the sample and charges 500 times by means of a shelf located inside the drum. The sample is then removed from the drum and sieved over a 1.70-mm (No. 12) sieve. The loss is calculated as the difference between initial mass and final mass. The loss is expressed to the nearest 1% of mass. For use in this project, procedure step 9.1.1 (AASHTO T 96) was followed which allows the user to waive the rinsing of the tested sample if the aggregate is essentially free of adherent coatings and dust. TxDOT personnel informed ICAR 507 personnel that the washing of tested samples changes the loss by a negligible percent. 5.6.2 Results The results of the L.A. Abrasion test proved to be very consistent. For this reason, only two samples were tested. If the variance was found to be greater than 1.0, a third sample was tested. The three samples were then compared. If an outlier did not exist, the three samples were averaged. If an outlier was found, it was discarded and the remaining two test results were averaged. The test data can be found in the Appendix D. Since the loss is a direct result of the amount of hand sieving at the end of the test procedure, it was found that some user variability could be removed by the same person testing both samples for a given source. This was due to the amount of time the sample was sieved over the 1.70-mm (No. 12) sieve following abrasion and impact in the L.A. machine.

68

5.7 5.7.1

Freezing and Thawing Test Procedures An unconfined test for the freezing and thawing of aggregate is provided in the

AASHTO specification, AASTHTO T 103 (Soundness of Aggregates by Freezing and Thawing) (AASHTO, 2000). However, deciding on an unconfined freeze-thaw test

method proved difficult. Wide variations in the use of AASHTO T 103 are allowed as no cooling rate or absolute minimum temperature is defined. Both of these variables have been shown to affect degradation due to freezing and thawing and the relationship of the results to field performance (Pigeon et al., 1985; Powers, 1955; Rogers et al., 1989). Moreover, personal communications with state agencies and testing laboratories revealed wide differences in the test method. Some reported freezing aggregate to 15 F (-9.5 C) mostly submerged then thawing with a 70 F forced air draft. Another reported vacuum saturating samples and freezing them suspended in plastic bags to a temperature of -15 F (-26.1 C), and still another reported vacuum saturating the aggregate and freezing them in metal pans. All of these methods are approximate variations of the AASHTO T 103 test, and as mentioned in the literature review in Chapter 2, the T 103 test has been shown to be unrealistically harsh. The Canadian Standards Association, however, has adopted an unconfined freezing and thawing test of aggregates that has been designed to model actual field conditions and maximize the relationship to field performance (Rogers et al., 1989). Research at the Ministry of Transpiration of Ontario determined the optimum cooling rate and minimum freezing temperature of unconfined freeze-thaw tests to maximize loss and relationship to field conditions. These observations are also in accordance with the observations of others as mentioned in the literature review. The optimum salt solution strength and the effect of the number of cycles were also determined. 69

All of these variables have been shown by MTO and others to significantly affect freeze-thaw durability (Pigeon et al., 1985; Powers, 1955; Rogers et al., 1989; Shakoor et al., 1982). The science behind the CSA specification has addressed these issues, whereas AASHTO T 103 and its variations have not. The CSA standard has also been

recommended by NCHRP (Hanna et al., 2003), and the AASHTO T 103 method has not. Moreover, the CSA test method can be completed with a fraction of the time and difficultly required by the T 103 method. Testing the aggregates for this research with some variation of T 103 would have required equipment not currently available. Due to the good correlation with field performance, the ease and quickness of the test, the NCHRP recommendation, and the overwhelming scientific support, the CSA A23.2-24A specification (Test Method for the Resistance of Unconfined Coarse Aggregate to Freezing and Thawing) was chosen for this research. The Canadian freeze-thaw test (LS-614) (Ontario Ministry of Transportation, 1989) was conducted in the following manner. After washing, oven drying, and sieving the aggregate, samples were prepared by hand sieving the material according to the following gradation: 3/4 in 1/2 in 1/2 in 3/8 in 3/8 in No. 4 1250 grams 1000 grams 500 grams

Each size fraction prepared was then individually placed in autoclavable mason jars. The samples were soaked in the jars for 24 hours in a 3% sodium chloride solution. After soaking, the solution was drained from the samples, and air tight lids were placed on the jars. The samples were cooled to a temperature of -18 C (0 F) for 16 hours overnight. They were then removed and allowed to thaw at room temperature for

approximately 8 hours. After the fifth cycle the jars were filled with water and rinsed 70

five times. Finally, the samples were oven dried to constant mass at 110 C 5 C and sieved over the original sieve sizes. The percent loss was calculated, and the final loss was determined by the weighted average of the percent loss of the three jars. For the purposes of conducting this test a blast freezer was purchased, and adjustments were made to control the freezing rate according to the optimal freezing rate as determined by MTO (Wuerpel, 1939). The freezing rate of the freezer was monitored over twelve practice runs to ensure consistent freezing, and fans were placed in the freezer chamber to ensure uniform freezing of all samples. Personnel were present in the afternoons to turn on the freezer for cooling and in the mornings to open the freezer doors for thawing. Although the samples were not removed from the chamber every morning, a high-powered box fan circulated air at room temperature through the chamber. The samples were then rotated as specified before freezing again that afternoon. remainder of the test was conducted exactly as stated in the specification and above.
30.0 25.0 20.0 15.0 10.0 Temp (C) 5.0 0.0 -5.0 -10.0 -15.0 -20.0 -25.0 2:15 PM TRJar11 TLJar11 MJar11 BRJar11 BLJar11 TLAir11 TRAir11

The

3:15 PM

4:15 PM Time

5:15 PM

6:15 PM

7:15 PM

Figure 5.7.1-1: Plot of Canadian Freeze-Thaw Cooling Rates 71

A sample from the Brownwood quarry in Texas was tested in each of the Canadian freeze-thaw tests. The precision requirements of these tests were subject to those outlined in the British Standard BS 812 (British Standards Institution, 1990). Figure 5.7.1-2 displays the results of the Brownwood control samples included within each run. The limits vary according to the ranges specified for weight loss ranges within the specifications and are shown on the graph as well. As the data illustrate, the results were
Freeze/Thaw Control Results
8

Percent Loss

Brownwood Control Top Limit Bottom Limit

0 0 1 2 3 4 5 Test Number 6 7 8 9 10

Figure 5.7.1-2:

Brownwood Control Samples

consistent and within limits. Additionally, over 12 practice tests showed that the freezer used for these tests could reliably reproduce freezing and thawing conditions, and the control samples tested were remarkably consistent making this type of precision statement feasible. A plot of the temperature during freezing is shown in Figure 5.7.1-1. 72

TLAir11 and TRAir11 are the air temperatures in two locations in the freezer, and the other curves plotted are the temperatures of samples in five different locations in the freezer. 5.7.2 Results The Canadian freeze-thaw test was conducted according to the method previously outlined. The results by source are listed in the Appendix E. As discussed in the following chapter, these values were used in conjunction with micro-Deval and magnesium sulfate soundness loss values to determine if correlations exist between these laboratory tests. In addition, these values were used to determine if a correlation exists between these laboratory tests and field performance. The British specification for the aggregate crushing value test was used as the precision statement for the Canadian freeze-thaw test. Each source was tested twice, and if an outlier appeared to exist another sample was tested. If the source of inconsistency could be identified as one size fraction or jar of the sample, that size fraction or jar alone was retested. The aggregate crushing value spec worked quite well for sources of

average to high loss. However, for low loss sources it was almost impossible to meet the British specification requirement. The 7% range on either side of the mean specified was so small that two very precise results of a low-loss source would not pass. In these cases, a judgment had to be made based on the coefficient of variation, and the 7% range to determine which low-loss aggregates would be retested. Importance was also placed on the expected change in the average if a third sample yielded exactly one of the previous values.

73

5.8

Aggregate Crushing Value Test The Aggregate Crushing Value (ACV) test (BS 812: Part 110) is a test used to

measure the crushing strength of an aggregate. While the specification used for comparison in the ICAR 507 project was from the British Standards, variations of this aggregate test are recognized in Australia and New Zealand where the test is thought of favorably and used for qualification purposes (Goldsworthy, 2004). The British version was chosen because no U.S. method (AASHTO or ASTM) exists for determining aggregate strength (Hanna et al., 2003). The ACV test was reported by Hanna et al. as being a reasonable approach for determining aggregate strength. 5.8.1 Test Procedures Due to limited use of the ACV test in the United States, appropriate test equipment is not readily accessible. The Ontario Ministry of Transportation loaned ICAR the equipment necessary to perform the ACV test. The equipment consists of an openended steel cylinder which confines the aggregate sample during testing, a steel base plate which the cylinder and aggregate sample rest, and a steel plunger for transferring the load from the testing machine to the sample. A steel measuring device was also included to ensure a constant volume of aggregate was used for each test run. The test procedure is fairly simple. Oven-dry aggregate is placed in the provided measuring device in a compacted manner. The sample of aggregate is then weighed and recorded as the initial mass. The sample is placed in the steel test cylinder in three equal lifts, tamping each lift twenty-five times with a steel rod with a hemispherical head. The plunger is placed on top of the sample, and a 400 kN (90,000 lb) force is applied uniformly over a ten-minute period. Once the specified load is reached, the load is removed and the sample is removed from the test cylinder. The samples were removed from the testing cylinders by turning the cylinder on its side and striking the cylinder with 74

a rubber mallet. Removing the sample from the test cylinder proved to be difficult for some types of angular aggregates, such as granites, while smooth river gravels were removed with ease. The tested sample is then passed over a 1.70-mm (No. 12) sieve with the retained material weighed and recorded as the final mass. Percent loss is found as the difference between the original and final masses divided by the original mass and recorded to the nearest one percent. 5.8.2 Results The aggregate crushing value results were generally found to be precise. Twice, the two results did not fall within the plus/minus 7% of the average test result. The largest difference was within plus/minus 13% of the average and the other unsatisfactory test results were within plus/minus 11% of the average. The test specification calls for two additional samples to be prepared and tested. The four test results are then to be averaged. 5.9 Aggregate Crushing Value (SSD) Test The Aggregate Crushing Value (SSD) test (variation of BS 812: Part 110) is a test used to measure the crushing strength of an aggregate in a saturated-surface dry condition. The principle is the same as for the micro-Deval test; some aggregates are negatively affected by the introduction of moisture. Comparison of the ACV and ACV (SSD) should provide a good correlation with aggregate performance according to Goldsworthy (2004). 5.9.1 Test Procedures The test setup is identical to the aggregate crushing value test. The difference is the soaking of aggregate in water prior to testing in the compression machine. A soaking time of fifteen hours was chosen based on the absorption and specific gravity

75

specification (AASHTO T 85). The aggregate was then surface dried and placed in the testing apparatus and subjected to a compressive load. Upon reaching the required load, the aggregate was removed from the steel test cylinder and placed in an oven set at 110C (230F). The sample remained in the oven until constant mass was achieved. Once the sample reached a constant mass, the sample was removed and sieved over a 1.70-mm (No. 12) sieve. The amount of sample retained on the 1.70-mm (No. 12) sieve was weighed and recorded. The percent loss was then calculated as the loss (initial mass minus the final mass) divided by the initial mass multiplied by 100. The percent loss was calculated to the nearest 1%. 5.9.2 Results The results for the saturated-surface dry aggregate crushing value test were less precise when compared to the aggregate crushing value test. Five of the forty-two samples did not fall within the specification limits set forth in the aggregate crushing value specification. This could be due to the fact that the saturated-surface dry aggregate crushing value test is slightly harsher and produces a slightly higher loss on average. Figure 5.9.2-1 shows aggregate crushing values plotted against aggregate crushing values (SSD). A forty-five degree line is also displayed which shows a higher number of samples above the line denoting the loss is higher in the aggregate crushing value (SSD) test.

76

50

Aggregate Crushing Value (SSD), % Loss

40

1
30

20

10

0 0 10 20 30 40 50

Aggregate Crushing Value, % Loss

Figure 5.9.2-1: 5.10 Absorption Test

Comparison of Aggregate Crushing Value Tests Results

The absorption of an aggregate is studied to develop a correlation between the field performance of aggregate and absorption combined with micro-Deval loss. Research performed with respect to correlating micro-Deval with other tests has found micro-Deval to correlate better with soundness tests than strength tests. It has also been shown that as absorption increases, loss in the sulfate soundness test increases as well (Mather (1947), Rogers et al. (1989), and Phillips et al. (2000). With both micro-Deval loss and absorption showing a correlation with magnesium sulfate soundness loss, there may exist a relationship between micro-Deval loss and absorption. Phillips et al. (2000) found this to be true, but the relationship was not as strong for micro-Deval loss and absorption as it was for magnesium sulfate 77

soundness loss and absorption. He also found that micro-Deval loss multiplied by absorption correlated better with magnesium sulfate soundness loss than either of the tests did individually. As a side note, work done by Marks and Dubberke (1982) has determined that absorption and field performance of aggregate do not correlate well. 5.10.1 Test Procedures The absorption specification (AASHTO T 85) requires the water temperature to be 23.0 1.7C (73.4 3F). This was reached via heaters and pumps, similar to those used in fish tanks. The heaters were sensitive enough to sustain the water within 0.3C ( 0.5F). In order for the water to be sustained at constant temperature via the heaters, the temperature of the room was kept slightly lower than the specified water temperature of the testing tank. The sample that was tested consisted of a grading with a 9.5-mm (3/8-inch) nominal maximum size. As mentioned previously, this was chosen for two reasons: The bulk of the micro-Deval test sample grading A consists of this size and the samples are for comparison only, not qualification. The samples were placed in pie tins during soaking. The pie tins had holes, 6.7 mm (1/4 inch) in diameter, drilled in them to allow water to flow freely in and out of the tins during placement and removal of the tins from a water tank. The holes were drilled smaller than the aggregate size to prevent loss of sample. The sample was placed in the soaking tank in an oven-dry condition and soaked for the required fifteen to nineteen hour time specification. The sample was then removed from the water and placed on a towel. The aggregate was moved around on the towel until the film of water on the surface of the aggregate particles was no longer visible. Care was taken not to over-dry the aggregate particles. The sample was then weighed, the 78

mass recorded as the saturated-surface-dry (SSD) weight of the aggregate sample. The final step was to place the sample in an oven until constant mass was reached. The constant mass was recorded as the oven-dry weight. These two recorded masses (ovendry test sample in air and SSD test sample in air) were used to calculate the absorption of the aggregate sources. 5.10.2 Results Results calculated by ICAR 507 differed slightly from some results that were provided by suppliers for a given aggregate source, namely due to ICAR 507 only testing 9.5-mm (3/8-inch) maximum nominal size aggregate. 5.11 Specific Gravity Test Specific gravity values are produced as a result of absorption testing. Little research correlating specific gravity values and micro-Deval test results has been performed to date. According to Phillips et al. (2000), specific gravity has no effect on micro-Deval loss. Landgren has also performed testing on the specific gravity and found the specific gravity of an aggregate will change as the soaking time is varied (1994). 5.11.1 Test Procedures The specific gravity specification (AASHTO T 85) is identical to the absorption specification (AASHTO T 85) with the exception of an additional measurement required for the mass of the test sample in water. The same sample was used for determination of both the absorption test and the specific gravity test for each source. The sample was placed in the soaking tank in an oven-dry condition and soaked for the required fifteen to nineteen hour time specification. The sample was removed from the soaking water and placed in the specified wire mesh basket. The basket and sample were placed in water, agitated to remove any trapped air from the sample, and 79

suspended from the scale. The mass in water was recorded on a data sheet. The sample was then removed from the water and placed on a towel. The aggregate was moved around on the towel until the film of water on the surface of the aggregate particles was no longer visible. Care was taken not to over-dry the aggregate particles. The sample was then weighed, the mass recorded as the saturated-surface-dry (SSD) weight of the aggregate sample. Lastly, the sample was placed in an oven until constant mass was reached. The constant mass was recorded as the oven-dry weight. The three recorded masses (oven-dry test sample in air, SSD test sample in air, and saturated test sample in water) were used to calculate the bulk specific gravity, bulk specific gravity (SSD), and apparent specific gravity. 5.11.2 Results Results calculated by ICAR 507 differed slightly from some results that were provided by suppliers for a given aggregate source, namely due to ICAR 507 only testing 9.5 mm (3/8 inch) maximum nominal size aggregate. 5.12 Flat and Elongated Test The flat and elongated test is included in the test suite to see if the shape of the aggregate affects micro-Deval loss. It is known that a flat or elongated particle will more than likely have more loss than a round particle in the micro-Deval test, when comparing samples of the same mineralogy, due to the potential of corners chipping off. It has been shown through research that particle shape can significantly affect the field performance of aggregates used in hot-mix asphalt or railroad ballast (Maupin, 1970; Boucher and Selig, 1987; Ahlrich, 1996; Kandhal and Parker, 1998; and Kuennen, 2001). According to Hanna et al. (2003), most state agencies measure the ratio of particle dimensions rather than measuring the percentage of flat and elongated particles. ICAR 80

507 measured the width to thickness, length to thickness, and length to width ratios to the nearest one half of a ratio. These three ratios were then used to come up with a single number that could be used for comparison purposes, a particle-shape factor. The particle-shape factor was developed by the researchers to include the three dimensions (thickness, width, and length). First, each ratio was normalized with respect to the smallest ratio, then the three normalized ratios were multiplied together to obtain a particle-shape factor. The particle-shape factor puts the ratios listed above into a single number which can be used for comparison with the micro-Deval test results. The particleshape factor will yield a value of 1.00 for a perfect sphere while giving a higher value for a flat or elongated particle and an even higher value for a flat and elongated particle. 5.12.1 Test Procedures A four-station proportional caliper device was purchased through Gilson Company. This device was equipped with two small light-gauge steel angles to aid in the proportioning of the aggregate particles. Figure 5.12.1-1 shows the proportional caliper device with the additional hardware attached. ICAR 507 decided to measure the aggregate particle dimensional ratios to the nearest half of a ratio. The as-purchased device allowed for some subjectivity in measuring the particles that was negated with the addition of the straight edges. This hardware was attached to aid the user in determining which ratio was most closely associated with the actual particle ratio. The angles were thin enough that their effects were negligible in introducing errors.

81

Figure 5.12.1-1:

Apparatus for Measuring the Shape Ratios for Aggregate Particles

The material chosen for this test was either 12.5-mm (1/2-inch) or a 9.5-mm (3/8inch) maximum nominal size material. The results for each size were compared for several aggregates and found to be nearly identical. The 12.5-mm (1/2-inch) maximum nominal size aggregate was preferred because the size of the particles was easier to handle. In some cases, not enough 12.5-mm (1/2-inch) material was present so 9.5-mm (3/8-inch) material was used. These cases were noted. A minimum of fifty particles was determined to be sufficient through testing of numerous sources. As each aggregate particle was tested, the results were entered into a spreadsheet. A plot was developed for each source with number of particles versus the average particle shape ratios. Generally, the effect of an additional aggregate particle was minimal on the average particle shape ratios after thirty particles but some sources required fifty particles. It was decided to test fifty particles for every source for uniformity. 82

5.12.2 Results The results of the flat and elongated test were used to come up with a single particle-shape factor. This factor was found to be fairly precise in a multiple-user study performed by the author. The factor is very time consuming to obtain and other means of obtaining the width to thickness, length to thickness, and length to width ratios should be investigated, possibly by an automated machine. 5.13 Percent Fractured Particles

5.13.1 Test Procedures The percent fractured particles test was conducted according to the ASTM D 5821-01 specification (Determining the Percentage of Fractured Particles in Coarse Aggregate) (ASTM, 2001). The material chosen for this test was either 12.5 mm (1/2 inch) or 9.5 mm (3/8 inch) maximum nominal size material. The 12.5 mm nominal size aggregate was preferred as it better represented the condition of the aggregate prior to crushing to produce smaller gradations. In some cases, not enough 12.5 mm material was present, so 9.5 mm material was used. As specified, the test sample must be large enough so that the largest particle is not more than 1% of the sample mass. Therefore, for 12.5 mm material a 500 g sample was used and for 9.5 mm material a 200 g sample was used. The various aggregates were evaluated for percent with one or more fractured faces and percent with two or more fractured faces. 5.13.2 Results The majority of tests resulted in 100% for both categories. This is to be expected as the only aggregates that will usually have values much below 100% are gravels which only make up approximately 20% of all the sources investigated for this project. 83

Chapter 6: Discussion and Analysis

6.1 Introduction

6.1.1 Analysis Methodology The results and correlations section for each application was divided into a discussion of all of the applicable data points for that application, followed by individual discussions of the applicable rock type subgroups for that application. Only rock type subgroups which contained enough data points to provide meaningful correlations were presented. For the comprehensive portion of the discussion, the results of all of the aggregate tests were presented first to establish the effectiveness of the individual tests to differentiate between good, fair, and poor performers using micro-Deval test alone. Next, two-test combinations involving micro-Deval were studied to determine whether or not improvements in classification could be obtained using two different test results. Finally, two-test combinations not involving micro-Deval were studied. Only the combinations that produced meaningful results upon visual inspection are discussed. summary of the results was provided. With each figure a table was provided which details the percentage of good, fair, and poor performers which are correctly categorized. Through visual inspection, it was determined to divide the aggregates into two groups one including only the good performers and the other including only the fair and poor performers. For the individual test results, if the good performers fell below the percent loss presented they were considered successfully categorized, and if the fair and poor performers were above the specified loss percentage, they were considered successfully categorized. The specific 85 Finally, a

gravity results worked in reverse as the better performers had higher specific gravity values. For the two-test combinations, a quadrant, which outlines the loss limits of the two tests, was drawn in each case. Various success rates were computed to evaluate the effectiveness of the tests in correctly classifying the aggregates. Table 6.1.1-1

summarizes the formulas used in Microsoft Excel 2003 to compute the success rates all of which were utilized to discuss and determine the effectiveness of two-test combinations. Table 6.1.1-1: Analysis Formula Definitions Type Formula

Quadrant Success Rate (Number of good points within the quadrant)/(Total number of points within the quadrant) Good Success Rate (Number of good points within the quadrant)/(Total number of good points) Fair Success Rate (Number of fair points outside the quadrant)/(Total number of fair points) Poor Success rate (Number of poor points outside the quadrant)/(Total number of poor points) Overall Success Rate [(Good Success Rate)*(Total number of good points)+(Fair Success Rate)*(Total number of fair points)+(Poor Success Rate)*(Total number of poor points)]/(Total number of points)

Additionally, multiple trials were carried out to obtain the best success rate percentages in each case by varying loss limits, and overall success rate, which represents the weighted average, was used to make the final judgment on how effectively the 86

aggregates were classified. In some cases, the good performers made up an extremely large percentage of the data points. In such situations, it was sometimes possible that the optimized solution occurred when all of good performers would lie below the loss limit, and hence the overall success rate was not meaningful. However, such cases were addressed on the individual basis as they occurred, and conclusions were still possible using other success rate definitions. 6.1.2 Rock Type Subgroups The basic knowledge that certain rock types were more suitable than others for use in construction was known and applied in Ontario, Canada by 1905 (Rogers, 1990). Therefore, mineralogy was used to break up the existing graphs into subgraphs, and analysis discussed in Section 6.1.1 was carried out on each one. 6.1.3 Climatic Regions Subgroups It is well known that temperature and precipitation have a significant effect on the performance and longevity of roadways and structures. What is not known is how these climatic conditions affect the ability to predict the performance of a particular aggregate in a given region through standard aggregate tests. For the purposes of this research project, a modified Thornthwaite index of the United States, as illustrated in Figure 6.1.3-1, was utilized (Desolminihae, Hudson, and Ricci, 1986). To include the Canadian provinces, the boundaries of the northern regions were extended directly north. In instances where a state or province was divided into two or more regions, it was placed into the region containing its largest portion. If the state or province was evenly divided, it was excluded from consideration. Only the hot-mix asphalt application was analyzed by climatic regions as portland cement concrete, base

87

course, and open-graded friction course did not have enough data points to draw meaningful conclusions.

Region I II III IV V VI Figure 6.1.3-1:

Characteristics Wet and no freeze Wet and freeze-thaw cycling Wet, hard freeze and spring thaw Dry and no freeze Dry and freeze-thaw cycling Dry, hard freeze and spring thaw

Climatic Regions of the United States (Desolminihae, Hudson, and Ricci)

88

6.2 6.2.1

Performance Analysis for Hot-Mix Asphalt Aggregates Individual Tests Literature review indicated that micro-Deval has proven to be a good predictor of

performance. To test this assertion, performance graphs were created and are included in the Appendix M, where the performance of aggregates was divided into three categories good, fair, and poor. The following sections will analyze the results by computing how well the micro-Deval test can separate the good performances from the fair and poor performers compared to other tests carried out during this research project. Both

qualitative and quantitative analysis will be carried out. Additionally, conclusions will be drawn on whether or not micro-Deval in combination with another test can improve the overall success rate of the prediction. Figure 6.2.1-1 represents the performance spread of all aggregates used in
40.0 35.0 30.0 Micro-Deval, % Loss 25.0 20.0 15.0 10.0 5.0 0.0
Poor Fair Perform ance Good

Figure 6.2.1-1:

Micro-Deval vs. Performance 89

hot-mix asphalt and tested using micro-Deval. As the plot demonstrates, all values lie below 40% micro-Deval loss. There is considerable scatter within each category.

However, higher densities are observed between 5 and 15% for good performers, 9 and 15% for fair performers, and 17 to 22% for poor performers. To test how well microDeval can separate good performers from fair and poor aggregates, a micro-Deval threshold value was varied in Microsoft Excel 2003, with correct percentage predictions computed for each trial value for good, fair, and poor aggregates. A weighted percentage of the three was computed as well to see how good the prediction is overall. Table 6.2.1-1 provides the key values to demonstrate how the overall percentage first

Table 6.2.1-1: Micro-Deval Success Rate

TRIAL NUMBER 1 2 3 4 5 6 7 8 9 10 MD Value 7% 8% 11% 12% 13% 14% 17% 18% 20% 21% GOOD 17% 31% 54% 56% 63% 67% 75% 77% 81% 83% FAIR 100% 100% 73% 73% 65% 54% 42% 38% 27% 23% POOR 95% 90% 90% 90% 90% 85% 70% 65% 40% 30% OVERALL 55% 61% 66% 67% 69% 67% 65% 64% 58% 56%

increases then decreases as the micro-Deval threshold value is increased. Thus, it can be concluded that micro-Deval alone can separate the good aggregates from the fair and poor sources with the maximum overall success rate of 69%. Magnesium sulfate testing was carried out in an attempt to measure the aggregates resistances to freeze-thaw cycles. Figure 6.2.1-2 demonstrates the results when performance is plotted against test values for all aggregates used in hot-mix asphalt. 90

80.0 Magnesium Sulfate Soundness, % Loss 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0

Poor

Fair Perform ance

Good

Figure 6.2.1-2:

Magnesium Sulfate Soundness vs. Performance

As the figure demonstrates, considerable scatter is present within each performance category. Such behavior is not unexpected since large variations in test results were observed within each source during magnesium sulfate testing. Generally, all good performers fall below the 40% loss mark with considerably higher point density in the 0 to 10% range. Fair performers are more widely spread out between losses of 0 and 70% with noticeably higher point density in the 0 to 7.5% range. Poor performers are even more widely spread out between losses of 0 and 70%, but the region with higher point density has moved up to approximately 10 to 15%. Table 6.2.1-2 provides a summary of quantitative analysis that was performed to assess the success rate of the test. As the table demonstrates, the highest overall success rate achieved is 64%, which is lower that that achieved by micro-Deval test alone.

91

Table 6.2.1-2: MSS Success Rate

TRIAL NUMBER 1 2 3 4 5 6 7 8 9 10 MSS Value 2% 4% 8% 10% 12% 16% 20% 24% 34% 40% GOOD 35% 52% 67% 77% 79% 88% 92% 96% 98% 100% FAIR 88% 73% 42% 27% 23% 23% 15% 8% 8% 4% POOR 90% 85% 75% 65% 60% 40% 30% 25% 15% 10% OVERALL 60% 64% 62% 61% 60% 61% 59% 58% 57% 56%

L.A. Abrasion values were then analyzed, and its performance graph is shown in Figure 6.2.1-3. Once again, considerable scatter is present within each category. Higher
40

35 LA Abrasion, % Loss

30

25 20

15

10

Poor

Fair Perform ance

Good

Figure 6.2.1-3:

L.A. Abrasion vs. Performance

densities are observed for fair performers between 19 and 28% and for poor performers between 22 and 23%. Once again, Microsoft Excel 2003 was utilized to test the entire 92

L.A. Abrasion loss range to observe how the overall success percentage changes as the threshold value is slowly increased. Key points are shown in Table 6.2.1-3 to

Table 6.2.1-3: LAA Success Rate

TRIAL NUMBER 1 2 3 4 5 6 7 8 9 10 LAA Value 12% 14% 19% 20% 22% 23% 24% 25% 28% 29% GOOD 4% 13% 31% 33% 46% 52% 52% 56% 63% 65% FAIR 100% 100% 81% 81% 73% 73% 58% 46% 31% 19% POOR 100% 95% 85% 85% 80% 70% 60% 60% 45% 45% OVERALL 49% 53% 55% 56% 60% 61% 55% 54% 51% 49%

demonstrate the progression behavior - the success rate is first observed to increase and then decrease. The maximum success percentage is observed at a value of 61% for the L.A. Abrasion loss of 23%. Thus, it can be concluded that the L.A. Abrasion test is less accurate in correctly separating the aggregates than the micro-Deval test. Literature review indicated that the Canadian freeze-thaw test should provide good prediction performance. Figure 6.2.1-4 provides the performance separation graph for all aggregates tested in this research project. It should be noted that all aggregates suffered relatively small losses, with a maximum of 15% suffered by a good-performing material. Scattering of data is observed for each performance category, but greater densities can be observed between 1 and 4% for good performers, 1.5 and 3.5% for fair performers, and 6 and 7.5% for poor performers.

93

16.0 14.0 Canadian Freeze-Thaw, % Loss 12.0 10.0 8.0 6.0 4.0 2.0 0.0
Poor Fair Perform ance Good

Figure 6.2.1-4:

Canadian Freeze-Thaw Soundness vs. Performance

To quantitatively assess the success rate of the tests prediction, Microsoft Excel 2003 was utilized as before by varying the loss value for the entire range in 1% increments. Table 6.2.1-4 shows the key points to demonstrate how the overall percentage

Table 6.2.1-4: CFT Success Rate

TRIAL NUMBER 1 2 3 4 5 6 7 8 9 10 CFT Value 1% 2% 3% 5% 6% 7% 8% 10% 11% 14% GOOD 6% 31% 56% 83% 87% 90% 90% 94% 96% 98% FAIR 100% 85% 58% 42% 35% 31% 27% 12% 0% 0% POOR 100% 90% 90% 50% 40% 25% 20% 15% 15% 5% OVERALL 50% 57% 63% 65% 63% 61% 59% 56% 54% 53%

94

first increases and then decreases as the test loss value was increased. The peak overall success rate is 65%, and it is attained at the Canadian freeze-thaw loss value of 5%. Hence, it can be concluded that while the Canadian freeze-thaw test is a better predictor of performance than the L.A. Abrasion test, it is nonetheless worse than the micro-Deval test. Aggregate crushing value test measures the strength of the material as it is subjected to compressive force in a confined environment. Figure 6.2.1-5 indicates that the crushing values for all sources lie between 10 and 35%. Good performers are uniformly spread out between 12 and 31%; fair performers lie between 14 and 30% with slightly higher point density observed in the 19 to 27.5% range; fair performers are scattered relatively uniformly between about 15 and 34%.
35 Aggregate Crushing Value, % Loss

30

25

20

15

10

Poor

Fair Perform ance

Good

Figure 6.2.1-5:

Aggregate Crushing Value vs. Performance

95

Quantitative analysis was performed on the data by varying the percent loss from 10 to 35% to determine the maximum overall success rate of the test to separate good performers from poor and fair aggregates. Table 6.2.1-5 outlines the key points to emphasize the fact that the overall success percentage first increases and then decreases.

Table 6.2.1-5: ACV Success Rate

TRIAL NUMBER 1 2 3 4 5 6 7 8 9 10 ACV 11% 12% 13% 16% 17% 18% 22% 23% 28% 29% GOOD 0% 4% 10% 27% 33% 38% 69% 75% 92% 94% FAIR 100% 100% 96% 85% 81% 81% 54% 27% 4% 4% POOR 100% 100% 100% 85% 80% 75% 40% 40% 15% 15% OVERALL 47% 49% 51% 54% 55% 57% 59% 55% 53% 54%

The maximum success rate is 59%, and it is reached at a loss value of 22%. Hence, it is clear that the aggregate crushing value test is not as accurate as the micro-Deval test. Aggregate crushing value (SSD) test is very similar to the previously-discussed crushing value test except for the fact that the aggregate pores are saturated with water prior to the material being subjected to compression. Figure 6.2.1-6 shows the spread of points by the performance criterion. As the figure indicates, all good performers lie between 9 and 45% with much higher density between 11 and 30%. Fair performers lie between 12 and 32% with higher density observed between 18 and 28%. Poor

performers show the greatest spread with values ranging between 0 and 50%, but higher density is observed between 15 and 23%.

96

60 Aggregate Crushing Value (SSD), % Loss

50 40

30

20 10

Poor

Fair Perform ance

Good

Figure 6.2.1-6:

Aggregate Crushing Value (SSD) vs. Performance

Table 6.2.1-6 shows ten points that demonstrate the fluctuations in the overall

Table 6.2.1-6: WCV Success Rate

TRIAL NUMBER 1 2 3 4 5 6 7 8 9 10 WCV 8% 13% 16% 17% 20% 23% 27% 32% 38% 51% GOOD 2% 4% 23% 29% 50% 67% 85% 94% 98% 100% FAIR 100% 100% 88% 88% 69% 46% 12% 0% 0% 0% POOR 100% 100% 100% 89% 63% 37% 21% 16% 16% 5% OVERALL 48% 49% 56% 57% 58% 55% 52% 53% 55% 54%

success percentage value as the loss percentage is gradually increased. The maximum is observed to be 58% and occurs for the loss value of 20.0%. Considering the maximum 97

micro-Deval success rate, it can be concluded that the micro-Deval test is a better predictor of performance. Absorption properties of aggregates influence the material behavior during the exposure to freeze-thaw cycling. Hence, it is necessary to look at the absorption versus performance graph for possible trends. As Figure 6.2.1-7 demonstrates, good performers are spread out throughout the range of 0 to 6% with high density between 0 and 2.2%.
6.0

5.0 4.0 Absorption

3.0

2.0 1.0

0.0

Poor

Fair Perform ance

Good

Figure 6.2.1-7:

Absorption vs. Performance

Fair performers are scattered between 0 and 3.8% with high point densities between 0 and 1% as well as 2.1 and 2.5%. Poor performers are spread out between 0 and 4.8% with slightly higher density towards the bottom of the spectrum. Table 6.2.1-7 displays ten key points to demonstrate the fluctuations in the overall prediction success rate. The maximum overall success percentage of 58% is achieved at an absorption capacity of 0.5%. However, this overall success rate is still lower than that of the micro-Deval test.

98

Table 6.2.1-7: Absorption Success Rate

TRIAL NUMBER 1 2 3 4 5 6 7 8 9 10 ABS 0.50% 1.00% 1.50% 2.00% 2.50% 3.00% 3.50% 4.00% 4.50% 5.50% GOOD 33% 62% 71% 85% 88% 88% 90% 96% 98% 98% FAIR 88% 35% 31% 23% 4% 4% 4% 0% 0% 0% POOR 84% 63% 47% 32% 26% 16% 16% 16% 16% 5% OVERALL 58% 55% 56% 57% 53% 51% 52% 54% 55% 53%

Three different specific gravities were computed from the measurements taken during specific gravity testing bulk specific gravity, saturated surface-dry specific gravity, as well as the apparent specific gravity. All three exhibit identical trends, and therefore, only bulk specific gravity graph is discussed below. All three graphs are included in Appendix M. Figure 6.2.1-8 demonstrates that most good performers lie between the bulk specific gravities of 2.25 and 3 with higher density observed between 2.5 and 2.75. Fair performers are scattered between the bulk specific gravities of 2.4 and 2.9 with higher density observed between 2.58 and 2.67. Poor performers are widely spread out between bulk specific gravity values of 2.2 and 2.9 with slightly higher density towards the middle of the spectrum.

99

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 2.10
Poor Fair Perform ance Good

Figure 6.2.1-8:

Specific Gravity (Bulk) vs. Performance

Table 6.2.1-8 demonstrates the trend of the computed overall success percentage

Table 6.2.1-8: SG (Bulk) Success Rate

TRIAL NUMBER 1 2 3 4 5 6 7 8 9 10 SG(BULK) 2.10% 2.20% 2.30% 2.40% 2.50% 2.60% 2.70% 2.80% 2.90% 3.10% GOOD 0% 0% 4% 6% 12% 29% 77% 90% 98% 100% FAIR 100% 100% 100% 96% 88% 58% 23% 4% 0% 0% POOR 100% 95% 79% 68% 58% 37% 5% 5% 5% 5% OVERALL 47% 46% 45% 43% 41% 38% 48% 50% 53% 54%

as the bulk specific gravity value is gradually increased.

The maximum success

percentage is 54%. During similar numerical analysis of saturated surface-dry specific 100

gravity and apparent specific gravity, maximum overall success value of 54% was observed as well, which is lower then the previously-computed micro-Deval overall success rating. Particle shape factor is a measure of angularity of an aggregate, a property that could have a considerable impact on the performance of the material in the field. Figure 6.2.1-9 displays the results when performance is plotted against the particle shape factor.
5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00
Poor Fair Perform ance Good

Figure 6.2.1-9:

Particle Shape Factor vs. Performance

As the figure indicates, good performers lie in the range of 1.5 to 4 with noticeably higher density in the lower spectrum values ranging between 1.5 and 2.5. Fair performers are more uniformly scattered between particle shape factors of 1.5 and 4.5 with slightly higher density between 1.75 and 2.25. The scatter is even higher for poor performers with values ranging between 1.75 and 5.25, and slightly higher density is apparent toward the bottom of the spectrum. Table 6.2.1-9 shows the calculations carried out to

101

quantitatively assess the effectiveness of the test at separating the aggregates. Overall maximum success rate is 62%, a value considerably lower than that of the micro-Deval. Table 6.2.1-9: PSF Success Rate
TRIAL NUMBER 1 2 3 4 5 6 7 8 9 10 PSF 1.00% 1.50% 2.00% 2.50% 3.00% 3.50% 4.00% 4.50% 5.00% 5.50% GOOD 0% 0% 40% 85% 90% 98% 100% 100% 100% 100% FAIR 100% 100% 73% 31% 12% 4% 4% 0% 0% 0% POOR 100% 90% 85% 45% 25% 15% 10% 10% 10% 5% OVERALL 47% 45% 58% 62% 56% 56% 56% 55% 55% 54%

Computations for percentages of fractured particles in each source were carried out during the research project. Data were recorded for two cases particles with one or more fractured faces and particles with two or more fractured faces. Figures 6.2.1-10 and 6.2.1-11 graphically summarize the findings. Furthermore, Tables M.1 and M.2 in the Appendix M provide the same information summarized in a table format. From Figure 6.2.1-10, it is clear that material in all three performance criteria was mostly fractured. Fair performers have the least scatter of values that lie between 70 and 100% while both fair and good performers have values ranging between 0 and 100% with higher densities at the top of the spectrums.

102

120

100

Percent Crushed (1+)

80

60

40

20

Poor

Fair Perform ance

Good

Figure 6.2.1-10:

Percent Crushed (1+) vs. Performance

Table 6.2.1-10 quantitatively assesses the effectiveness of the test. The maximum overall success percentage is 54%, a value significantly smaller than that of the micro-Deval test.

Table 6.2.1-10:
TRIAL NUMBER 1 2 3 4 5 6 7 8 9 10

Percent Crushed (1+) Success Rate

GOOD 4% 4% 4% 4% 4% 4% 8% 8% 10% 100% FAIR 96% 96% 96% 92% 92% 92% 88% 81% 77% 0% POOR 100% 100% 100% 100% 100% 100% 95% 70% 75% 5% OVERALL 48% 48% 48% 47% 47% 47% 47% 40% 41% 54%

%Crushed (1+) 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 100.00%

103

Figure 6.2.1-11 provides the graphical summary for the test where particles with two or more fractured faces were counted. The results are similar to those of the one fractured face analysis. However, greater scatters are observed for poor performers, which now lie between 60 and 100%.
120

100

Percent Crushed (2+)

80

60

40

20

Poor

Fair Perform ance

Good

Figure 6.2.1-11:

Percent Crushed (2+) vs. Performance

Both good and fair performers are scattered throughout the entire range of 0 to 100% with slightly higher point concentrations around 100%. Table 6.2.1-11 provides

quantitative analysis summary. The observed maximum overall success percentage is 54% as in the previous case of only one fractured face analysis. Thus, it can be

concluded that the percent fractured test is not as good of a performance predictor as the micro-Deval test regardless whether one or two fractured faces are counted.

104

Table 6.2.1-11:
TRIAL NUMBER 1 2 3 4 5 6 7 8 9 10

Percent Crushed (2+) Success Rate

GOOD 4% 4% 4% 4% 4% 4% 8% 10% 15% 100% FAIR 96% 92% 92% 92% 92% 88% 85% 81% 77% 0% POOR 100% 100% 100% 100% 100% 95% 95% 70% 75% 5% OVERALL 48% 47% 47% 47% 47% 45% 46% 41% 44% 54%

%Crushed (2+) 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 100.00%

6.2.2

Combinations Involving Micro-Deval One of the main goals of this research project was to identify whether or not

micro-Deval in combination with another test would be able to predict performance better than if it was used alone. Literature review suggested that a quadrant could be identified on a graph where the majority of good performers would lie. Therefore, analysis will be carried out as described in Section 6.1. Figure 6.2.2-1 represents the results when the micro-Deval test is plotted against the L.A. Abrasion test. From the plot, it can be observed that good performers tend to congregate towards the lower-left corner.

105

70 60 LA Abrasion, % Loss 50 40 30 20 10 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure 6.2.2-1:

L.A. Abrasion vs. Micro-Deval

To quantitatively assess the additional information gained from the L.A. Abrasion test, numerous trials were carried out by changing the position of the borders of the quadrant. Table 6.2.2-1 summarizes the general trends observed when the borders were moved to the right or to the top, with Trial 3 being the optimal quadrant position obtained through extensive trial and error process. In the table, the value for quadrant success percentage is computed by calculating the number of good points within the quadrant and dividing it by the total number of points within the quadrant. The value of good percentage is

computed by dividing the number of good points within the quadrant by the total number of good points. Similarly, fair percentage is computed by dividing the number of fair points outside the quadrant by the total number of fair performers, and poor percentage is computed by dividing the number of poor performers outside the quadrant by the total number of poor performers. Finally, overall success percentage is the weighted average of good, poor, and fair percentages. 106

Table 6.2.2-1: LAA and MD Success Rate

TRIAL NUMBER LAA MD QUADRANT GOOD FAIR POOR OVERALL 1 18 10 100% 17% 100% 100% 56% 2 23 13.5 88% 42% 92% 95% 66% 3 52 13.5 74% 65% 62% 90% 69%

Hence, the table values provide a measure of successful aggregate qualification for a twotest combination, where higher quadrant and overall percentages indicate better aggregate qualification performance. Thus, Table 6.2.2-1 shows that an overall success rate of 69% can be achieved but at the expense of the quadrant percentage. Furthermore, as the quadrant is expanded, the percentage of good aggregates accurately predicted increases significantly while the percentages for the fair and poor aggregates decrease. Therefore, it can be concluded that combining micro-Deval with L.A. Abrasion test does not yield better results than using the micro-Deval test alone, which can achieve 69% success rate on its own as discussed in the previous section. The literature review indicated that micro-Deval and Canadian freeze-thaw tests combination provides good results for aggregate separation. Since magnesium sulfate soundness test is intended to measure aggregates resistance to freeze-thaw, similar results were expected. Figure 6.2.2-2 demonstrates the results of micro-Deval and

magnesium sulfate soundness being included on the same plot.

107

40.0 Magnesium Sulfate Soundness, % Loss

POOR
35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

FAIR GOOD

Figure 6.2.2-2:

Magnesium Sulfate Soundness vs. Micro-Deval

As the figure demonstrates, considerable scatter was observed. However, good performers seem to congregate in the lower-left quadrant. Once again, quantitative analysis was performed, and the findings are summarized in Table 6.2.2-2.

Table 6.2.2-2: MSS and MD Success Rate

TRIAL NUMBER MSS MD QUADRANT GOOD FAIR POOR OVERALL 1 7 5 100% 12% 100% 100% 53% 2 8 8.5 94% 31% 100% 95% 62% 3 17 12.5 77% 63% 69% 90% 70%

108

Three trials are outlined to provide an overview of quadrant expansion progression in an attempt to increase the overall success rate. As the quadrant is expanded further to the right and to the top, quadrant success rate decreases. However, the overall success rate of 70% becomes possible. This is a slight improvement upon the 69% overall success rate that can be achieved by micro-Deval alone. As previous discussion indicated, the Canadian freeze-thaw test is one of the better tests to separate good performers from fair and poor aggregates. Furthermore, the literature review indicated that the combination of micro-Deval and Canadian freezethaw tests produced better predictions than using each test alone. Figure 6.2.2-3 displays the results when micro-Deval is plotted against Canadian freeze-thaw. The three

quadrants outlined in the plot represent the results of numerous trials in an attempt to optimize overall prediction success percentage. During the first attempt, outlined by the
16.0

POOR
14.0 Canadian Freeze-Thaw, % Loss 12.0 10.0 8.0 6.0 4.0 2.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

FAIR GOOD

Figure 6.2.2-3:

Canadian Freeze-Thaw vs. Micro-Deval

109

solid line, only good performers were included in the quadrant, resulting in an overall success rate of 63%. By including several fair sources into the quadrant area, an overall success rate of 69% can be achieved. By further extending the border to the right, the overall percentage increases to the maximum value of 73%. Numerical results for the three trials are summarized in Table 6.2.2-3. As the table indicates, the overall success rate

Table 6.2.2-3: CFT and MD Success Rate

TRIAL NUMBER CFT MD QUADRANT GOOD FAIR POOR OVERALL 1 5 8 100% 31% 100% 100% 63% 2 5 10 89% 48% 88% 100% 69% 3 5 19 75% 75% 62% 85% 73%

is achieved at the expense of decreasing fair and poor prediction success rates as well as the quadrant prediction success rate. Thus, it can be concluded that the additional

information provided by the Canadian freeze-thaw test is quite significant as its inclusion in the analysis leads to the higher overall success rate than that of using the micro-Deval test alone. Figure 6.2.2-4 represents the results when micro-Deval losses are plotted against aggregate crushing value losses.

110

50 Aggregate Crushing Value, % Loss 45 40 35 30 25 20 15 10 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure 6.2.2-4:

Aggregate Crushing Value vs. Micro-Deval

For the first trial, the quadrant was chosen so that only good performers would fall into the area. Overall success rate of 53% is achieved in this case but only 12% of good aggregates overall are classified correctly. The quadrant was then expanded by slowly increasing the micro-Deval loss value. The newly-enclosed area now contains several poor sources, but the overall success rate increases to 59% with the improvement of good performer prediction to 27%. By further expanding the quadrant to the right and upwards, the overall success rate increases to 69%. During this process, the quadrant prediction percentage falls significantly as does the percentage for fair prediction. However, 65% of good performers now fall into the quadrant. Thus, it can be concluded that aggregate crushing value test does not provide any useful additional information to increase the overall success rate since micro-Deval alone produces a 69% success rate.

111

Table 6.2.2-4: ACV and MD Success Rate

TRIAL NUMBER ACV MD QUADRANT GOOD FAIR POOR OVERALL 1 20 5 100% 12% 100% 100% 53% 2 20 9.5 88% 27% 100% 90% 59% 3 32 13.5 74% 65% 62% 90% 69%

Since the aggregate crushing value (SSD) test is very similar to the aggregate crushing value test, similar results were expected and observed during the data analysis. Figure 6.2.2-5 shows the plotted results with three curves outlining the outcomes as the
55 Aggregate Crushing Value (SSD), % Loss 50 45 40 35 30 25 20 15 10 5 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.2.2-5:

Aggregate Crushing Value (SSD) vs. Micro-Deval

quadrant borders were expanded. Table 6.2.2-5 provides numerical summary for the trials data. During the first trial, only good performers are included in the quadrant area. 112

Even though 59% overall success rate is achieved, only 12% of good sources overall are counted. As the quadrant is allowed to expand to the right, two poor sources are included in the quadrant, hence reducing its success rating to 88%. However, more good

Table 6.2.2-5: WCV and MD Success Rate

TRIAL NUMBER WCV MD QUADRANT GOOD FAIR POOR OVERALL 1 22 5 100% 12% 100% 100% 53% 2 33 8.5 88% 31% 100% 89% 61% 3 33 13.5 73% 65% 62% 89% 69%

performers overall are now included (31% of all good aggregates), and the overall success rate increases to 61%. Upon further quadrant expansion, the quadrant success rate continues to decline to 73% as more fair performers now fall into it, and fair performance success rate falls significantly to 62%. Overall success rate increases to 69%, however, and 65% of good performers now fall into the quadrant. Numerous trials were carried out during quantitative analysis, and 69% shown in Trial #3 is the maximum achieved overall success rate. Thus, it can be concluded that saturated, surface-dry aggregate crushing value test produces results very similar to those of the regular crushing value test, and hence no significant improvements are made to the overall success rate of 69% obtained through the use of the micro-Deval test alone. Figure 6.2.2-6 represents the results obtained by plotting micro-Deval losses and absorption capacities of the aggregates on the same plot. assessment information is summarized in Table 6.2.2-6. Additionally, quadrant

113

6.0

5.0

4.0 Absorption

3.0

2.0

POOR
1.0

FAIR GOOD

0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

Figure 6.2.2-6:

Absorption vs. Micro-Deval

The first trial represents the results obtained by including only good performers within the quadrant, resulting in 52% success rate, but only 10% of all good points fall within the quadrant. By expanding the quadrant boundary to the right, both poor sources success rate and the quadrant success rate decrease,

Table 6.2.2-6: ABS and MD Success Rate

TRIAL NUMBER ABS MD QUADRANT GOOD FAIR POOR OVERALL 1 2 5 100% 10% 100% 100% 52% 2 2 7.5 92% 23% 100% 95% 58% 3 2.2 12.5 78% 62% 73% 89% 70%

114

but the percent of good points falling within the area increases to 92% and the overall success rate improves to 58%. By further expanding the area to the right and slightly upward, the maximum overall success rate of 70% is observed. However, as expected, since more fair sources now fall into the quadrant, its success rate as well as fair aggregates success rates fall significantly. Thus, it can be concluded that the addition of absorption data to the micro-Deval data increases the prediction success rate very slightly to 70% as compared to 69% achieved by micro-Deval alone. Based on the data collected during the research project, three specific gravity values were computed bulk specific gravity, saturated surface-dry specific gravity, and apparent specific gravity. Only bulk specific gravity plotted against micro-Deval is discussed here since the data for all three forms the exact same scatter patterns, but the readers is encouraged to refer to the Appendix M for the specific gravity graphs. Figure 6.2.2-7 shows the results after numerous trials were carried out aimed at optimizing the overall success rate.
3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.2.2-7:

Specific Gravity (Bulk) vs. Micro-Deval 115

Table 6.2.2-7 provides the numerical summary of the trials data. During the first trial, only good performers are included in the quadrant, resulting in 52% success rate, but only 10% of good performers fall into the area. Upon expansion of the quadrant to the right, several poor points fall into the quadrant area, resulting in quadrant success rate decreasing to 88%, but the overall success rate increases to 60%, and now 29% of good

Table 6.2.2-7: SG(Bulk) and MD Success Rate

TRIAL NUMBER SG(BULK) MD QUADRANT GOOD FAIR POOR OVERALL 1 3 5 100% 10% 100% 100% 52% 2 3 8.5 88% 29% 100% 89% 60% 3 3.1 13 74% 62% 65% 89% 68%

performers fall into the square. The maximum success rate of 68% is observed during Trial #3 but at the expense of fair sources success rate decreasing from 100% to 65% and quadrant success rate falling significantly to 74%. Thus, it is clear that the addition of specific gravity information does not improve our predictions results at all compared to using the micro-Deval test alone. Particle shape factor provides information about the degree of angularity of aggregate particles. Figure 6.2.2-8 demonstrates the final results after numerous trials were carried out in an attempt to optimize the overall success rate. Table 6.2.2-8

provides numerical summary for the key steps in the trial process. The first trial again is intended to have the quadrant contain good performers only.

116

5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.2.2-8:

Particle Shape Factor vs. Micro-Deval

The overall success rate of 53% is achieved in this case, but only 12% of all good performers are included in the quadrant area. By expanding the quadrant to the right, the overall success rate increases to 59% with 25% of good performers falling into the quadrant area. Further gains are possible at the expense of quadrant success rate decrease by expanding the area. Trial #3 represents the optimal quadrant size to achieve the

Table 6.2.2-8: PSF and MD Success Rate

TRIAL NUMBER PSF MD QUADRANT GOOD FAIR POOR OVERALL 1 2.5 5 100% 12% 100% 100% 53% 2 2.6 7.5 93% 25% 100% 95% 59% 3 4 12.5 77% 63% 69% 90% 70%

117

highest possible overall success rate of 70%. This result leads to the inclusion of 63% of good performers into the quadrant area but significant decreases in quadrant success rate as well as fair sources success rate. Thus, it can be concluded that very slight

improvement to prediction success rate is observed by including particle shape factor information to that of the micro-Deval test alone. Figure M.24 and Figure M.25 in Appendix M demonstrates the results of plotting micro-Deval losses and percent crushed results on the same plot. As the graphs indicate, no clear data groupings are evident, and hence no further statistical analysis is possible. Thus, it can be concluded that percent crushed tests, one crushed face or two crushed faces, do not provide any additional information. 6.2.3 Other Relevant Combinations During this research project, a series of two-dimensional graphs was created where every test was plotted against every other test. This series of graphs is included in the Appendix M. Every graph was carefully examined in an attempt to find out whether or not meaningful quadrants could be drawn in as was done in the micro-Deval case. The results of this inspection lead to the conclusion that only two other test combinations produced meaningful quadrant divisions: Canadian freeze-thaw versus aggregate

crushing value graph produced and magnesium sulfate soundness versus particle shape factor. Both graphs are discussed in this section. Figure 6.2.3-1 shows the plot of aggregate crushing value versus Canadian freezethaw losses. Once again, a statistical analysis was performed as before to evaluate how well the good aggregates could be separated. Results of this analysis are summarized in Table 6.2.3-1.

118

50

POOR
Aggregate Crushing Value, % Loss 45 40 35 30 25 20 15 10 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 Canadian Freeze-Thaw , % Loss

FAIR GOOD

Figure 6.2.3-1:

Aggregate Crushing Value vs. Canadian Freeze-Thaw

As the graph data show, higher overall success rate was achieved at the expense of decreasing quadrant success rate. While the decreases in success rates for fair and poor performers are not very significant, they do increase significantly in order to achieve the maximum overall success rate of 67%. However, the increase in the success rate of good performers is quite significant to justify the decreases in fair and poor success rates.

Table 6.2.3-1: ACV and CFT Success Rate

TRIAL NUMBER ACV CFT QUADRANT GOOD FAIR POOR OVERALL 1 25 1.9 92% 21% 96% 100% 57% 2 28 1.9 83% 29% 92% 95% 59% 3 28 4.4 67% 75% 46% 75% 67%

119

Generally, however, the maximum overall success rate achieved by aggregate crushing value and Canadian freeze-thaw tests is not as good as that achieved through the use of the micro-Deval test in combination with other tests. Figure 6.2.3-2 provides the graph of particle shape factor versus magnesium sulfate soundness.
5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure 6.2.3-2:

Particle Shape Factor vs. Canadian Freeze-Thaw

The quantitative analysis performed on this graph is summarized in Table 6.2.3-2 below.

Table 6.2.3-2: PSF and MSS Success Rate

TRIAL NUMBER PSF MSS QUADRANT GOOD FAIR POOR OVERALL 1 2.4 4.5 89% 48% 92% 95% 69% 2 2.4 8.5 81% 65% 77% 90% 73%

120

As the table demonstrates, by expanding the quadrant slightly to the right and to the top the quadrant success rate decreases by 8%. However, the overall success rate increases to 73% - a notable improvement to the success rate of 69% achieved by micro-Deval test alone. Thus, it can once again be concluded that freeze-thaw performance evaluation provides significant information to successful identify aggregate performance in the field. 6.2.4 Results Summary Table 6.2.4-1 below provides the summary of the statistical analysis performed for hot-mix asphalt aggregates. The first column represents the success rates obtained when only one test is used to separate the good-performing aggregates. The second column provides the success rate of a two-test combination with one of the tests being

Table 6.2.4-1: HMA Success Rates Summary

MICRO-DEVAL MAGNESIUM SULFATE SOUNDNESS L.A. ABRASION CANADIAN FREEZE-THAW AGGREGATE CRUSHING VALUE AGGREGATE CRUSHING VALUE (SSD) ABSORPTION SPECIFIC GRAVITY (BULK) PARTICLE SHAPE FACTOR % CRUSHED (1+) % CRUSHED (2+) TEST ALONE 69% 64% 61% 65% 59% 58% 58% 54% 58% 54% 54% MICRO-DEVAL COMBINATION NOT APPLICABLE 70% 69% 73% 69% 69% 70% 68% 70% NOT APPLICABLE NOT APPLICABLE

the micro-Deval test. As the data considering one test only illustrate, the highest success rate percentage was achieved using the micro-Deval test, which was followed by the Canadian freeze-thaw test. These conclusions are supportive of the conclusions drawn by previous research projects outlined in the literature review section. The data in the 121

second column indicate that success rates could be improved from the highest of 69% by using micro-Deval alone to 73% overall success rate by using micro-Deval and Canadian freeze-thaw tests together. This result is in line with research performed by Chris Rogers, et al, who concluded that better classification results are obtained by combining Canadian freeze-thaw test data with micro-Deval test data. Thus, it is the conclusion of this research project team that micro-Deval represents the best single-test prediction performance while the combination of micro-Deval and Canadian freeze-thaw tests represents the best two-test combination for performance classification of aggregates used in the hot-mix asphalt. 6.2.5 Limestone and Dolomite 6.2.5.1 Combinations Involving Micro-Deval For the limestone and dolomite subgroup to hot-mix asphalt, micro-Deval showed a clear separation between poor performers and the good and fair performers. From Figure 6.2.5.1-1 and Table 6.2.5.1-1, all of the poor performers had a loss value greater than 18.8%. However, there appeared to be no difference between fair and good

performers, and 27% of the good performers had loss values above 18.8%. This value of 18% separating the poor performers correlates with Kandhal and Parkers work in 1998 (Kandhal and Parker, 1998).

122

40.0 35.0 30.0 Micro-Deval, % Loss 25.0 20.0 15.0 10.0 5.0 0.0

Poor

Fair
Performance

Good

Figure 6.2.5.1-1:

Micro-Deval vs. Performance

Table 6.2.5.1-1:
Trial 1 2 3 4 5 6 7 8 9 10 MD 7% 8% 11% 12% 13% 14% 17% 18.8% 20% 40% Good 0% 0% 13% 13% 33% 33% 53% 73% 73% 100%

MD Success Rate
Fair 100% 100% 88% 88% 88% 63% 38% 38% 25% 0% Poor 100% 100% 100% 100% 100% 100% 100% 100% 63% 0% Overall 52% 52% 55% 55% 65% 58% 61% 71% 58% 48%

123

The combination of microDeval and did magnesium not yield sulfate good

Table 6.2.5.1-2:
Trial MSS MD Quadrant Good Fair Poor Overall 1 7% 5% N/A N/A N/A N/A N/A

MSS vs. MD Success Rate

2 8% 8.5% N/A N/A N/A N/A N/A 3 17% 12.5% 83% 33% 88% 100% 65% 4 6% 15% 88% 47% 88% 100% 71%

soundness

correlations with field performance in this subgroup. From Figure 6.2.5.1-2 and Table 6.2.5.1-2, the bounds from the comprehensive section enclosed

very few points. The best correlation existed at trial 4 (denoted by the blue lines) which had an optimized solution of 71%. This was no better than micro-Deval alone.
40.0 Magnesium Sulfate Soundness, % Loss 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.2.5.1-2:

Magnesium Sulfate Soundness vs. Micro-Deval

124

From Figure 6.2.5.1-3 and Table 6.2.5.13, no correlations existed when micro-Deval was combined with L.A. abrasion. The

Table 6.2.5.1-3: LAA vs. MD Success Rate

Trial LAA MD Quadrant Good Fair Poor Overall 1 18% 10% N/A N/A N/A N/A N/A 2 23% 13.5% 80% 27% 88% 100% 61% 3 52% 13.5% 71% 33% 75% 100% 61%

optimized solution was an 18.8% micro-Deval loss alone as illustrated by the blue line.

60

50 LA Abrasion, % Loss

40

30

20

10

POOR FAIR GOOD

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0

0 Micro-Deval, % Loss

Figure 6.2.5.1-3:

L.A. Abrasion vs. Micro-Deval

125

Micro-Deval in combination with Canadian Freeze-Thaw provided the most meaningful correlations. From Figure 6.2.5.1-4 and Table 6.2.5.1-4, the bounds from the

Table 6.2.5.1-4:
Trial CFT MD Quadrant Good Fair Poor Overall 1 5% 8% N/A N/A N/A N/A N/A

CFT vs. MD Success Rate

2 5% 10% 67% 13% 88% 100% 55% 3 5% 19% 78% 47% 88% 88% 68% 4 6% 18.6% 89% 53% 88% 100% 74%

comprehensive section (refer to trial 3) which represented good limits for

good performers. The best overall success rate was represented by the blue lines which successfully categorized 74% of the total.
16.0 14.0 Canadian Freeze-Thaw, % Loss 12.0 10.0 8.0 6.0 4.0 2.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.2.5.1-4:

Canadian Freeze-Thaw vs. Micro-Deval

126

Combining aggregate crushing value with micro-Deval showed little correlation for this subgroup. From Figure 6.2.5.1-5 and Table 6.2.5.1-5, the best solution was once again at a micro-Deval loss of 18.8% as represented by the blue line.

Table 6.2.5.1-5: ACV vs. MD Success Rate

Trial ACV MD Quadrant Good Fair Poor Overall 1 20% 5% N/A N/A N/A N/A N/A 2 20% 9.5% 100% 7% 100% 100% 55% 3 32% 13.5% 71% 33% 75% 100% 61%

60

Aggregate Crushing Value, % Loss

50

40

30

20

10

POOR FAIR GOOD

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0

0 Micro-Deval, % Loss

Figure 6.2.5.1-5:

Aggregate Crushing Value vs. Micro-Deval

127

As was illustrated in Figure 6.2.5.1-6 and Table 6.2.5.1-6, the results for aggregate crushing value (SSD) and micro-Deval were very similar to the oven dry aggregate crushing value and micro-Deval results. The optimized solution was the 18.8% micro-Deval loss as represented by the blue line.

Table 6.2.5.1-6: WCV vs. MD Success Rate

Trial WCV MD Quadrant Good Fair Poor Overall 1 22% 5% N/A N/A N/A N/A N/A 2 33% 8.5% N/A N/A N/A N/A N/A 3 33% 13.5% 71% 33% 75% 100% 61%

60 Aggregate Crushing Value (SSD), % Loss

50

40

30

20

10

POOR FAIR GOOD

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0

0 Micro-Deval, % Loss

Figure 6.2.5.1-6:

Aggregate Crushing Value (SSD) vs. Micro-Deval

128

The bounds detailed in the comprehensive section for absorption and micro-Deval covered only a few of the points for limestone and dolomite. From Figure 6.2.5.1-7 and Table 6.2.5.17, it was obvious this combination had little correlation with field performance.

Table 6.2.5.1-7:
Trial ABS MD Quadrant Good Fair Poor Overall

ABS vs. MD Success Rate

1 2% 5% N/A N/A N/A N/A N/A 2 2% 7.5% N/A N/A N/A N/A N/A 3 2.2% 12.5% 83% 33% 88% 100% 65%

The best result was once again the micro-Deval test alone.
5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.2.5.1-7:

Absorption vs. Micro-Deval

129

No strong correlations existed for this subgroup with the combination of absorption and micro-Deval. As was shown in Figure

Table 6.2.5.1-8: SG(Bulk) vs. MD Success Rate

Trial SG(Bulk) MD Quadrant Good Fair Poor Overall 1 2.6 5% N/A N/A N/A N/A N/A 2 2.6 8.5% N/A N/A N/A N/A N/A 3 2.6 13% 83% 33% 88% 100% 65%

6.2.5.1-8, the optimized solution was the microDeval test alone.

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.2.5.1-8:

Specific Gravity (Bulk) vs. Micro-Deval Table 6.2.5.1-9: PSF vs. MD Success Rate
Trial PSF MD Quadrant Good Fair Poor Overall 1 2.5 5% N/A N/A N/A N/A N/A 2 2.6 7.5% N/A N/A N/A N/A N/A 3 4 12.5% 83% 33% 88% 100% 65%

The combination of particle shape factor with micro-Deval proved no better than most of the previous combinations. From Figure 6.2.5.1-9 and Table 6.2.5.1-9, the optimized solution was the microDeval limit of 18.8% alone.

130

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.2.5.1-9:

Particle Shape Factor vs. Micro-Deval

6.2.5.2 Other Relevant Combinations From Figure 6.2.5.2-1 and Table 6.2.5.2-1, the combination of aggregate crushing value and Table 6.2.5.2-1:
Trial ACV CFT Quadrant Good Fair Poor Overall

ACV vs. CFT Success Rate

1 25% 1.9% N/A N/A N/A N/A N/A 2 28% 1.9% 0% 0% 100% 88% 48% 3 28% 4.4% 54% 47% 75% 50% 55% 4 42% 6% 63% 67% 75% 50% 65%

Canadian freeze-thaw resulted in an optimized solution of only 65%. These bounds were also very bad at isolating the poor performers,

therefore this correlation from the comprehensive section did not exist in this subgroup.

131

60

Aggregate Crushing Value, % Loss

50

40

30

20

10

0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

16.0

Figure 6.2.5.2-1:

Aggregate Crushing Value vs. Canadian Freeze-Thaw

6.2.5.3 Section Results Summary Similar to the comprehensive section, the test combination with the best correlation to field performance for limestone and dolomite was micro-Deval and Canadian freeze-thaw. Table 6.2.5.3-1 illustrated the combination of these two tests increased the success rate from 71% for micro-Deval alone to 74%. From Table 6.2.5.14, the bounds specified for this combination only successfully qualified 53% of the good performers, so they were not good for eliminating potential sources. However, if a source fits within these bounds it would most likely perform well.

132

Table 6.2.5.3-1:
Test

HMA Limestone and Dolomite Success Rate Summary

Test Alone 71% 65% 61% 61% 68% 57% 61% 58% 71% Micro-Deval Combination N/A 71% 71% 74% 71% 71% 71% 71% 71%

Micro-Deval Magnesium Sulfate Soundness L.A. Abrasion Canadian Freeze-Thaw Aggregate Crushing Value Aggregate Crushing Value (SSD) Absorption Specific Gravity (Bulk) Particle Shape Factor

6.2.6

Siliceous Gravel

6.2.6.1 Combination Involving Micro-Deval For the siliceous gravel subgroup of hot-mix asphalt, the micro-Deval test results better isolated the good performers from the fair and poor performers. From Figure 6.2.6.1-1 and Table 6.2.6.1-1, all of the good performers had micro-Deval losses less than 11%. This value also eliminated 67% of the poor and fair performers. The three fair and poor performers below the 11% line were from northern locations which experienced more severe freezing and thawing. The best overall success rate achieved was 79%.

133

40.0 35.0 30.0 Micro-Deval, % Loss 25.0 20.0 15.0 10.0 5.0 0.0

Poor

Fair Performance

Good

Figure 6.2.6.1-1:

Micro-Deval vs. Performance

Table 6.2.6.1-1:
Trial 1 2 3 4 5 6 7 8 9 10 MD 7% 8% 11% 12% 13% 14% 17% 18% 20% 30% Good 20% 40% 100% 100% 100% 100% 100% 100% 100% 100%

MD Success Rate
Fair 100% 100% 60% 60% 60% 60% 60% 60% 60% 0% Poor 100% 75% 75% 75% 75% 50% 25% 25% 0% 0% Overall 71% 71% 79% 79% 79% 71% 64% 64% 57% 36%

134

The combination of micro-Deval and magnesium sulfate soundness did not yield good correlations with field performance in this subgroup. 6.2.6.1-2, From Figure 6.2.6.1-2 and Table the limits specified in the

Table 6.2.6.1-2: MSS vs. MD Success Rate

Trial MSS MD Quadrant Good Fair Poor Overall 1 7% 5% 100% 20% 100% 100% 71% 2 8% 8.5% 100% 40% 100% 100% 79% 3 17% 12.5% 63% 100% 60% 75% 79%

comprehensive section resulted in a best case success rate of 79%. This was exactly the same as micro-Deval alone as indicated by the blue line.
25.0 Magnesium Sulfate Soundness, % Loss

20.0

15.0

10.0

5.0

POOR FAIR GOOD

0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

0.0

Figure 6.2.6.1-2:

Magnesium Sulfate Soundness vs. Micro-Deval

135

From Figure 6.2.6.1-3 and Table 6.2.6.13, no correlations existed when micro-Deval was combined with L.A. abrasion. The

Table 6.2.6.1-3: LAA vs. MD Success Rate

Trial LAA MD Quadrant Good Fair Poor Overall 1 18% 10% 100% 20% 100% 100% 71% 2 23% 13.5% 60% 60% 80% 75% 71% 3 52% 13.5% 63% 100% 60% 75% 79%

optimized solution was an 11% micro-Deval loss alone as illustrated by the blue line.

60

50 LA Abrasion, % Loss

40

30

20

10

POOR FAIR GOOD

0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

Figure 6.2.6.1-3:

L.A. Abrasion vs. Micro-Deval

136

Micro-Deval in combination with Canadian freeze-thaw provided one of the more meaningful

Table 6.2.6.1-4:
Trial CFT MD Quadrant Good Fair Poor Overall

CFT vs. MD Success Rate

2 5% 10% 80% 80% 80% 100% 86% 3 5% 19% 71% 100% 80% 75% 86% 4 4% 11% 83% 100% 80% 100% 93%

correlations for the siliceous gravel subgroup. From Figure 6.2.6.1-4 and Table 6.2.6.1-4, the bounds from the comprehensive section were effective

1 5% 8% 100% 40% 100% 100% 79%

in isolating the good performers (refer to trials 1, 2, and 3). The best possible solution was detailed in trial 4 and illustrated with the blue line. These limits produced an overall success rate of 93%. The lone fair point within these bounds, designated source #114, had mixed performance results. In dry environments it performed well, but in wet environments it performed poorly.
12.0

10.0 Canadian Freeze-Thaw, % Loss

8.0

6.0

4.0

2.0

POOR FAIR GOOD

0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

0.0

Figure 6.2.6.1-4:

Canadian Freeze-Thaw vs. Micro-Deval

137

Combining aggregate crushing value with micro-Deval improved the correlation some for this subgroup. From Figure 6.2.6.1-5 and Table 6.2.6.1-5, setting the limits at 20% and 10%, respectively, produced an

Table 6.2.6.1-5:
Trial ACV MD Quadrant Good Fair Poor Overall 1 20% 5% 100% 20% 100% 100% 71%

ACV vs. MD Success Rate

2 20% 9.5% 75% 60% 100% 75% 79% 3 32% 13.5% 63% 100% 60% 75% 79% 4 20% 10% 80% 80% 100% 75% 86%

overall success rate 7% higher than micro-Deval alone.

35 30 25 20 15 10 5 0 0.0 5.0 10.0

Aggregate Crushing Value, % Loss

POOR FAIR GOOD

15.0 Micro-Deval, % Loss 20.0 25.0 30.0

Figure 6.2.6.1-5:

Aggregate Crushing Value vs. Micro-Deval

138

As

illustrated

in

Figure

Table 6.2.6.1-6:
Trial WCV MD Quadrant Good Fair Poor Overall 1 22% 5% 100% 20% 100% 100% 71%

6.2.6.1-6 and Table 6.2.6.1-6, the results for aggregate crushing value (SSD) and micro-Deval provided very good correlations with performance. The limits described in trial 4 and illustrated by the blue line

WCV vs. MD Success Rate

2 33% 8.5% 67% 40% 100% 75% 71% 3 33% 13.5% 63% 100% 60% 75% 79% 4 17% 10% 100% 80% 100% 100% 93%

successfully categorized 93% of the aggregates, an increase of 14% over micro-Deval alone.
35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

POOR FAIR GOOD

Figure 6.2.6.1-6:

Aggregate Crushing Value (SSD) vs. Micro-Deval

139

The

bounds

detailed

in

the

Table 6.2.6.1-7:
Trial ABS MD Quadrant Good Fair Poor Overall

comprehensive section for absorption and micro-Deval eliminated most of the poor and fair performers, but their overall success is no better than micro-Deval alone. From Figure 6.2.6.1-7 and Table 6.2.6.1-7, the optimized success rate was 79%.
4.0 3.5 3.0 Absorption, % 2.5 2.0 1.5 1.0 0.5 0.0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0

ABS vs. MD Success Rate

1 2% 5% 100% 20% 100% 100% 71% 2 2% 7.5% 100% 40% 100% 100% 79% 3 2.2% 12.5% 63% 100% 60% 75% 79%

POOR FAIR GOOD

25.0 30.0

Figure 6.2.6.1-7:

Absorption vs. Micro-Deval

140

No strong correlations existed for this subgroup with the combination of specific gravity and micro-Deval. As was indicated in Figure 6.2.6.1-8 and Table 6.2.6.1-8, the optimized solution was the micro-Deval test alone at a loss limit of 11%.

Table 6.2.6.1-8: SG(Bulk) vs. MD Success Rate

Trial SG(Bulk) MD Quadrant Good Fair Poor Overall 1 2.5 5% 100% 20% 100% 100% 71% 2 2.5 8.5% 67% 40% 100% 75% 71% 3 2.5 13% 63% 100% 60% 75% 79%

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

POOR FAIR GOOD

30.0

Figure 6.2.6.1-8:

Specific Gravity (Bulk) vs. Micro-Deval

141

The combination of particle shape factor and micro-Deval proved surprisingly effective for this

Table 6.2.6.1-9:
Trial PSF MD Quadrant Good Fair Poor Overall 1 2.5 5% 100% 20% 100% 100% 71%

PSF vs. MD Success Rate

2 2.6 7.5% 100% 40% 100% 100% 79% 3 4 12.5% 63% 100% 60% 75% 79% 4 2 11% 100% 80% 100% 100% 93%

subgroup. From Figure 6.2.6.1-9 and Table 6.2.6.1-9, the optimized solution resulted in an overall success rate of 93%.

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

POOR FAIR GOOD

Figure 6.2.6.1-9:

Particle Shape Factor vs. Micro-Deval

142

6.2.6.2 Other Relevant Combinations From Figure 6.2.6.2-1 and Table 6.2.6.2-1, the combination of aggregate crushing value and Table 6.2.6.2-1:
Trial ACV CFT Quadrant Good Fair Poor Overall

ACV vs. CFT Success Rate

2 28% 1.9% 100% 20% 100% 100% 71% 3 28% 4.4% 71% 100% 60% 100% 86% 4 22% 4% 83% 100% 80% 100% 93%

Canadian freeze-thaw resulted in a high overall success rate of 93%. From the figure, it was obvious Canadian freeze-thaw was a very

1 25% 1.9% 100% 20% 100% 100% 71%

effective test at eliminating poor performers within this subgroup.

30

Aggregate Crushing Value, % Loss

25

20

15

10

0 0.0 2.0 4.0 6.0 8.0 10.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

12.0

Figure 6.2.6.2-1:

Aggregate Crushing Value vs. Canadian Freeze-Thaw

143

6.2.6.3 Section Results Summary For the siliceous gravel subgroup of hot-mix asphalt, several combinations of tests provided good correlations with field performance. As Table 6.2.6.3-1 indicated, the best combinations were micro-Deval with Canadian freeze-thaw, aggregate crushing value (SSD), and particle shape factor and Canadian freeze-thaw with aggregate crushing value. The small amount of data points compared to the previous section increased the relative weight each data point carried. Based on this, it was difficult to draw

conclusions, but it appeared as though the performance of siliceous gravels could be more accurately predicted with lab tests than most aggregate types.

Test Micro-Deval Magnesium Sulfate Soundness L.A. Abrasion Canadian Freeze-Thaw Aggregate Crushing Value Aggregate Crushing Value (SSD) Absorption Specific Gravity (Bulk) Particle Shape Factor

Test Alone 79% 79% 64% 86% 79% 79% 57% 57% 79%

Micro-Deval Combination N/A 79% 79% 93% 86% 93% 79% 79% 93%

144

Table 6.2.6.3-1:

HMA Siliceous Gravel Success Rate Summary

6.2.7

Granite

6.2.7.1 Combinations Involving Micro-Deval The granite subgroup of hot-mix asphalt contained only fair and good performers. From Figure 6.2.7.1-1 and Table 6.2.7.1-1, a clear delineation existed between the two categories at 8% micro-Deval loss. Only one data point, a good performer designated source #99, fell on the wrong side of the line resulting in an overall success rate of 92%; this represented a 25% improvement over the total percentage of good performers. None of the test combinations were able to isolate this one wayward point. Thus, the overall success rate did not go above 92%.

145

40.0 35.0 30.0 Micro-Deval, % Loss 25.0 20.0 15.0 10.0 5.0 0.0

Poor

Fair Performance

Good

Figure 6.2.7.1-1:

Micro-Deval vs. Performance

Table 6.2.7.1-1:
Trial 1 2 3 4 5 6 7 8 9 10 MD 7% 8% 11% 12% 13% 14% 17% 18% 20% 21%

MD Success Rate
Good 38% 88% 100% 100% 100% 100% 100% 100% 100% 100% Fair 100% 100% 25% 25% 0% 0% 0% 0% 0% 0% Overall 58% 92% 75% 75% 67% 67% 67% 67% 67% 67%

146

The combination of micro-Deval and magnesium sulfate soundness did not improve the success rate for this subgroup. From Figure 6.2.7.1-2 and Table 6.2.7.1-2, the bounds detailed in trial 2 from the comprehensive section provided a success rate as high as microDeval alone as indicated by the blue line.
18.0 Magnesium Sulfate Soundness, % Loss 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 0.0 2.0 4.0 6.0 8.0

Table 6.2.7.1-2: MSS vs. MD Success Rate

Trial MSS MD Quadrant Good Fair Overall 1 7% 5% 100% 25% 100% 50% 2 8% 8.5% 100% 88% 100% 92% 3 17% 12.5% 73% 100% 25% 75%

FAIR GOOD
10.0 12.0 14.0 Micro-Deval, % Loss

Figure 6.2.7.1-2:

Magnesium Sulfate Soundness vs. Micro-Deval

147

From Figure 6.2.7.1-3 and Table 6.2.7.13, no correlations existed when micro-Deval was combined with L.A. abrasion. The

Table 6.2.7.1-3: LAA vs. MD Success Rate

Trial LAA MD Quadrant Good Fair Overall 1 18% 10% 100% 13% 100% 42% 2 23% 13.5% 100% 25% 100% 50% 3 52% 13.5% 67% 100% 0% 67%

optimized solution was an 8% micro-Deval loss alone as illustrated by the blue line.

60

50 LA Abrasion, % Loss

40

30

20

10

FAIR GOOD
0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0

0 Micro-Deval, % Loss

Figure 6.2.7.1-3:

L.A. Abrasion vs. Micro-Deval

Micro-Deval in combination with Canadian Freeze-Thaw did not improve the overall success rate either. From

Table 6.2.7.1-4: CFT vs. MD Success Rate

Trial CFT MD Quadrant Good Fair Overall 1 5% 8% 100% 88% 100% 92% 2 5% 10% 88% 88% 75% 83% 3 5% 19% 67% 100% 0% 67%

Figure 6.2.7.1-4 and Table 6.2.7.1-4, the bounds from the comprehensive section given in trial 1 provided the optimized

148

success rate of 92%.

However, this was due to the optimized success rate occurring at a

micro-Deval loss of 8% (refer to blue line).

6

5 Canadian Freeze-Thaw, % Loss

FAIR GOOD
0 2 4 6 8 10 12 14 16 18 20

0 Micro-Deval, % Loss

Figure 6.2.7.1-4:

Canadian Freeze-Thaw vs. Micro-Deval

Combining aggregate crushing value with micro-Deval did not improve the

Table 6.2.7.1-5: ACV vs. MD Success Rate

Trial ACV MD Quadrant Good Fair Overall 1 20% 5% 100% 25% 100% 50% 2 20% 9.5% 100% 25% 100% 50% 3 32% 13.5% 67% 100% 0% 67%

correlation for this subgroup.

From Figure

6.2.7.1-5 and Table 6.2.7.1-5, the optimized solution occurred at a micro-Deval loss of 8%.

149

35 30 Aggregate Crushing Value, % Loss 25 20 15 10 5 0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 Micro-Deval, % Loss

FAIR GOOD

Figure 6.2.7.1-5:

Aggregate Crushing Value vs. Micro-Deval

As was illustrated in Figure 6.2.7.1-6 and Table 6.2.7.1-6, the results for aggregate crushing value (SSD) and micro-Deval provided no improvement in overall success rate. The limits described in trial 2 and illustrated by the blue line successfully categorized 92% of the

Table 6.2.7.1-6: WCV vs. MD Success Rate

Trial WCV MD Quadrant Good Fair Overall 1 22% 5% 100% 25% 100% 50% 2 33% 8.5% 100% 88% 100% 92% 3 33% 13.5% 67% 100% 0% 67%

aggregates. This was the same as micro-Deval alone.

150

35 Aggregate Crushing Value (SSD), % Loss 30

25 20 15 10

5 0 0.0 2.0 4.0 6.0 8.0 10.0 12.0

FAIR GOOD
14.0 16.0

Micro-Deval, % Loss

Figure 6.2.7.1-6:

Aggregate Crushing Value (SSD) vs. Micro-Deval

No strong correlations existed for this subgroup with the combination of absorption and micro-Deval. As was

Table 6.2.7.1-7:
Trial ABS MD Quadrant Good Fair Overall

ABS vs. MD Success Rate

1 2% 5% 100% 25% 100% 50% 2 2% 7.5% 100% 63% 100% 75% 3 2.2% 12.5% 73% 100% 25% 75%

shown in Figure 6.2.7.1-7 and Table 6.2.7.1-7, the optimized solution was the micro-Deval test alone at a loss limit of 8%.

151

2.5

2.0

Absorption, %

1.5

1.0

0.5

FAIR GOOD
0.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Micro-Deval, % Loss

Figure 6.2.7.1-7:

Absorption vs. Micro-Deval

From Figure 6.2.7.1-8 and Table 6.2.7.18, the limits for specific gravity and microDeval utilized in the comprehensive section, as described in trial 2, provided an optimized success rate equal to micro-Deval alone.

Table 6.2.7.1-8: SG(Bulk) vs. MD Success Rate

Trial SG(Bulk) MD Quadrant Good Fair Overall 1 2.5 5% 100% 25% 100% 50% 2 2.5 8.5% 100% 88% 100% 92% 3 2.5 13% 67% 100% 0% 67%

152

3.1 3 2.9 Specific Gravity (Bulk) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 0 2 4 6 8 10 12 14 Micro-Deval, % Loss

FAIR GOOD

Figure 6.2.7.1-8:

Specific Gravity (Bulk) vs. Micro-Deval

The combination of particle shape factor and micro-Deval proved no better than micro-Deval alone. 6.2.7.1-9 and Table From Figure 6.2.7.1-9, the

Table 6.2.7.1-9:
Trial PSF MD Quadrant Good Fair Overall

PSF vs. MD Success Rate

1 2.5 5% 100% 25% 100% 50% 2 2.6 7.5% 100% 63% 100% 75% 3 4.0 12.5% 73% 100% 25% 75%

optimized solution occurred at a microDeval loss of 8% as indicated by the blue line.

153

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Micro-Deval, % Loss

FAIR GOOD

Figure 6.2.7.1-9:

Particle Shape Factor vs. Micro-Deval

6.2.7.2 Other Relevant Combinations From Figure 6.2.7.2-1 and Table 6.2.7.21, the combination of aggregate crushing value and Canadian freeze-thaw resulted in a high overall success rate of 83%; 9% less than microDeval alone. Table 6.2.7.2-1: ACV vs. CFT Success Rate
Trial ACV CFT Quadrant Good Fair Overall 1 25% 1.9% 100% 38% 100% 58% 2 28% 1.9% 100% 75% 100% 83% 3 28% 4.4% 70% 88% 25% 67%

154

35 30 25 20 15 10 5 0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 Canadian Freeze-Thaw, % Loss

Aggregate Crushing Value, % Loss

FAIR GOOD

Figure 6.2.7.2-1:

Aggregate Crushing Value vs. Canadian Freeze-Thaw

6.2.7.3 Section Results Summary For the granite subgroup of hot-mix asphalt, the most effective correlations were provided by the micro-Deval test alone. As Table 6.2.7.3-1 illustrated, no combination of tests was able to improve the success rate above 92%. One good performing aggregate, source #99, could not be isolated from the fair performers by any combination of tests. Table 6.2.7.3-1:
Test Micro-Deval Magnesium Sulfate Soundness L.A. Abrasion Canadian Freeze-Thaw Aggregate Crushing Value Aggregate Crushing Value (SSD) Absorption Specific Gravity (Bulk) Particle Shape Factor

HMA Granite Success Rate Summary

Test Alone 92% 83% 67% 75% 75% 75% 42% 67% 83% Micro-Deval Combination N/A 92% 92% 92% 92% 92% 92% 92% 92%

155

6.2.8

Climatic Regions

6.2.8.1 Region II The region II results indicated for a

Trial 1 2 3 4 5 6 7

Table 6.2.8.1-1:
MD 5% 10% 13% 15% 20% 25% 40% Good 9% 41% 55% 64% 77% 91% 100%

MD Success Rate
Fair 100% 86% 79% 57% 29% 7% 0% Poor 100% 100% 100% 80% 40% 40% 0% Overall 51% 63% 68% 63% 56% 56% 54%

maximum success rate with micro-Deval of 68% at a loss of 13% as indicated by trial 3 of Table 6.2.8.1-1.

As is shown in Figure 6.2.8.1-1, the points are fairly well scattered with the good performers mostly lying below a 20% loss.
45.0 40.0 35.0 Micro-Deval, % Loss 30.0 25.0 20.0 15.0 10.0 5.0 0.0
Poor Fair Perform ance Good

Figure 6.2.8.1-1:

Micro-Deval vs. Performance

From Table 6.2.8.1-2, micro-Deval provided the highest success rate of all the tests. In addition, no combination of tests improved upon the success rate from microDeval alone. 156

Table 6.2.8.1-2:
Test

HMA Region II Success Rate Summary

Test Alone 68% 59% 61% 63% 61% 59% 56% 56% 66% Micro-Deval Combination N/A 68% 68% 68% 68% 68% 68% 68% 68%

Micro-Deval Magnesium Sulfate Soundness L.A. Abrasion Canadian Freeze-Thaw Aggregate Crushing Value Aggregate Crushing Value (SSD) Absorption Specific Gravity (Bulk) Particle Shape Factor

Based on previous findings difficulty concerning in the

Trial 1 2 3 4 5 6 7

Table 6.2.8.1-3:
MD 5% 10% 12% 15% 20% 25% 35% Good 14% 64% 79% 79% 79% 93% 100%

MD Success Rate
Fair 100% 86% 71% 71% 29% 14% 0% Poor 100% 100% 100% 67% 33% 33% 0% Overall 50% 75% 79% 75% 58% 63% 58%

correctly

predicting the performance of limestone and dolomite, these rock types were

removed from the data set for region II. As is illustrated in Figure 6.2.8.1-3 and Table 6.2.8.1-3, this significantly improved the success rate for region II. While the percentage of good performers went up 4%, the overall percentage for micro-Deval went up 11%. This further substantiated the belief that it is very difficult to accurately predict the performance of limestone and dolomite.

157

40.0 35.0 30.0 Micro-Deval, % Loss 25.0 20.0 15.0 10.0 5.0 0.0
Poor Fair Perform ance Good

Figure 6.2.8.1-3:

Micro-Deval vs. Performance

Table 6.2.8.1-4 indicates that even with the limestones and dolomites removed, the overall success rate for micro-Deval could not be improved by combining it with other tests. Table 6.2.8.1-4:
Test Micro-Deval Magnesium Sulfate Soundness L.A. Abrasion Canadian Freeze-Thaw Aggregate Crushing Value Aggregate Crushing Value (SSD) Absorption Specific Gravity (Bulk) Particle Shape Factor

HMA Region II No Limestone/Dolomite Success Rate Summary

Test Alone 79% 67% 67% 71% 63% 67% 67% 71% 67% Micro-Deval Combination N/A 79% 79% 79% 79% 79% 79% 79% 79%

158

6.2.8.2 Region III The results for

Trial 1 2 3 4 5 6

Table 6.2.8.2-1:
MD 5% 10% 13% 15% 20% 35% Good 8% 75% 92% 92% 92% 100%

MD Success Rate
Fair 100% 100% 83% 67% 33% 0% Poor 100% 100% 100% 86% 29% 0% Overall 56% 88% 92% 84% 60% 48%

region III showed very good correlations between field performance and test results. From Figure 6.2.8.2-1 and Table 6.2.8.2-1, the

optimized success rate for micro-Deval occurred at a loss value of 13%. With an overall percentage of 92%, this loss value represented a 54% improvement over the percentage of good performers.
40.0 35.0 30.0 Micro-Deval, % Loss 25.0 20.0 15.0 10.0 5.0 0.0
Poor Fair Perform ance Good

Figure 6.2.8.2-1:

Micro-Deval vs. Performance

159

Several combinations of tested improved the overall success rate for hot-mix asphalt in region III to 96%. The first of these combinations

Table 6.2.8.2-2:
Trial LAA MD Quadrant Good Fair Poor Overall 1 18% 10% 100% 25% 100% 100% 64%

LAA vs. MD Success Rate

2 23% 13.5% 100% 42% 100% 100% 72% 3 52% 13.5% 92% 92% 83% 100% 92% 4 40% 13% 100% 92% 100% 100% 96%

involved L.A. Abrasion and microDeval, as is illustrated in Figure 6.2.8.2-2 and Table 6.2.8.2-2. The

blue lines in Figure indicate a L.A. Abrasion loss limit of 40% and micro-Deval loss limit of 13%.
60

50 LA Abrasion, % Loss

40

30

20

POOR
10

FAIR GOOD
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0

0 Micro-Deval, % Loss

Figure 6.2.8.2-2:

L.A. Abrasion vs. Micro-Deval

160

Figure 6.2.8.2-3 illustrates the combination of aggregate crushing value and micro-Deval. As is

Table 6.2.8.2-3:
Trial ACV MD Quadrant Good Fair Poor Overall 1 20% 5% 100% 8% 100% 100% 56%

ACV vs. MD Success Rate

2 20% 9.5% 100% 33% 100% 100% 68% 3 32% 13.5% 92% 92% 83% 100% 92% 4 28% 13% 100% 92% 100% 100% 96%

illustrated by the blue lines and trial 4 of Table 6.2.8.2-3, this combination provided an overall success rate of 96% as well.

35 30 25 20 15 10

Aggregate Crushing Value, % Loss

POOR
5 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 Micro-Deval, % Loss

FAIR GOOD
35.0

Figure 6.2.8.2-3:

Aggregate Crushing Value vs. Micro-Deval

161

The final combination which provided an increased overall success rate was aggregate crushing value (SSD) and micro-Deval. The results for these two tests are plotted in Figure 6.2.8.2-4. The blue lines Table 6.2.8.2-4:
Trial WCV MD Quadrant Good Fair Poor Overall 1 22% 5% 100% 8% 100% 100% 56%

WCV vs. MD Success Rate

2 33% 8.5% 100% 42% 100% 100% 72% 3 33% 13.5% 92% 92% 83% 100% 92% 4 29% 13% 100% 92% 100% 100% 96%

indicate the results of trial 4 from Table 6.2.8.2-4. As is shown the

overall success rate is 96%.

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10

POOR
5 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 Micro-Deval, % Loss

FAIR GOOD
35.0

Figure 6.2.8.2-4:

Aggregate Crushing Value (SSD) vs. Micro-Deval

The results for region III indicated very good correlations between field performance and aggregate tests. As is illustrated in Table 6.2.8.2-5, micro-Deval alone gave the best overall success rate of any individual test. In addition, the combination of micro-Deval with L.A. abrasion, aggregate crushing value, or aggregate crushing value (SSD) provided near perfect characterization. 162

Table 6.2.8.2-5:
Test

HMA Region III Success Rate Summary

Test Alone 92% 88% 60% 80% 56% 56% 80% 72% 56% Micro-Deval Combination N/A 92% 96% 92% 96% 96% 92% 92% 92%

Micro-Deval Magnesium Sulfate Soundness L.A. Abrasion Canadian Freeze-Thaw Aggregate Crushing Value Aggregate Crushing Value (SSD) Absorption Specific Gravity (Bulk) Particle Shape Factor

6.2.8.3 Region V The data set for region V consisted of 16 aggregates of which 12 were good performers and 4 were poor performers. Due to this high percentage of good performers, all of the overall success rates were relatively high. The results for micro-Deval are indicated in Figure 6.2.8.3-1 and Table 6.2.8.3-1. The optimized Table 6.2.8.3-1:
Trial 1 2 3 4 5 MD 5% 10% 15% 20% 30% Good 8% 33% 83% 83% 100%

MD Success Rate
Poor 100% 75% 75% 50% 0% Overall 31% 44% 81% 75% 75%

solution occurred at a loss value of 15% and had an overall success rate

only 6% higher than the percentage of good performers.

163

40.0 35.0 30.0 Micro-Deval, % Loss 25.0 20.0 15.0 10.0 5.0 0.0

Poor

Fair Perform ance

Good

Figure 6.2.8.3-1: Micro-Deval vs. Performance From Table 6.2.8.3-2, the results for region V did not show very good correlations with field performance and aggregate tests. Two tests, magnesium sulfate soundness and aggregate crushing value (SSD), provided higher overall percentages than micro-Deval, however, this was because each of these data sets had one less poor performer. Table 6.2.8.3-2:
Test Micro-Deval Magnesium Sulfate Soundness L.A. Abrasion Canadian Freeze-Thaw Aggregate Crushing Value Aggregate Crushing Value (SSD) Absorption Specific Gravity (Bulk) Particle Shape Factor

HMA Region V Success Rate Summary

Test Alone 81% 87% 75% 75% 81% 87% 80% 80% 81% Micro-Deval Combination N/A 87% 81% 81% 81% 87% 80% 80% 81%

164

6.2.8.4 Region VI The results for region VI provided very high percentages, but this was mainly due to the small number of data points. This data set consisted of 3 good performers, 4 fair performers, and 3 poor performers. The largest percentage of without regard to

optimization was represented by the total percentage of fair and poor performers and resulted in a 70% success rate. As Table 6.2.8.4-1 indicates, solution the for optimized micro-deval
Trial 1 2 3 4 5 6

Table 6.2.8.4-1:
MD 0% 5% 10% 15% 20% 25% Good 0% 67% 67% 67% 67% 100%

MD Success Rate
Fair 100% 100% 100% 25% 25% 0% Poor 100% 100% 67% 67% 33% 0% Overall 70% 90% 80% 50% 40% 30%

occurred at a loss limit of 5%. This limit only

effectively isolated 67% of the good performers, but due to the large percentage of poor and fair performers it resulted in a overall success rate of 90%. The results are graphed in Figure 6.2.8.4-1.
40.0 35.0 30.0 Micro-Deval, % Loss 25.0 20.0 15.0 10.0 5.0 0.0

Poor

Fair Perform ance

Good

Figure 6.2.8.4-1:

Micro-Deval vs. Performance 165

The results for region VI provided very high overall success rates, but this was mainly due to limited data set. From Table 6.2.8.4-2, micro-Deval, magnesium sulfate soundness, and aggregate crushing value (SSD) provided the highest overall success rates. In addition, no combination of tests improved the success rate beyond 90%. Table 6.2.8.4-2:
Test Micro-Deval Magnesium Sulfate Soundness L.A. Abrasion Canadian Freeze-Thaw Aggregate Crushing Value Aggregate Crushing Value (SSD) Absorption Specific Gravity (Bulk) Particle Shape Factor

HMA Region VI Success Rate Summary

Test Alone 90% 90% 80% 80% 70% 90% 80% 70% 80% Micro-Deval Combination N/A 90% 90% 90% 90% 90% 90% 90% 90%

166

6.3 6.3.1

Performance Analysis for Portland Cement Concrete Aggregates Individual Tests In 1998, Kandhal and Parker (Kandhal and Parker, 1998) reported that a micro-

Deval loss limit of 18% appeared to separate good and fair performing aggregates from poor performing aggregates in hot-mix asphalt. For this study, Figure 6.3.1-1 and Table 6.3.1-1 illustrates a micro-Deval loss limit of 21% represented the optimized solution for overall successful qualification in portland cement concrete. While the majority of good performers had a loss below 21%, so did the majority of poor and fair performers. The optimized solution, as noted in trial 5, is only 2% higher than the total percentage of good performers as indicated in trial 8.
40.0 35.0 30.0 Micro-Deval, % Loss 25.0 20.0 15.0 10.0 5.0 0.0

Poor

Fair
Performance

Good

Figure 6.3.1-1:

Micro-Deval vs. Performance

167

Table 6.3.1-1: MD Success Rate

Trial 1 2 3 4 5 6 7 8 MD 5% 10% 15% 20% 21% 25% 30% 35% Good 13% 59% 74% 87% 95% 95% 95% 100% Fair 75% 50% 50% 50% 50% 25% 0% 0% Poor 100% 80% 60% 20% 20% 0% 0% 0% Overall 27% 60% 71% 77% 83% 79% 77% 81%

As stated in the literature review, some researchers have suggested that poor performing aggregates have magnesium sulfate soundness losses greater than 18% to 20% (Kandhal and Parker, 1998; Paul, 1932). As illustrated in Figure 6.3.1-2, very few aggregates tested had losses higher than 18%. However, of these 60% were good

performing aggregates. Due to the wide range of all three performance levels, the optimized solution (refer to trial 5 of Table 6.3.1-2) had an overall success rate no greater than the total percentage of good performers as indicated in trial 8.
35.0 Magnesium Sulfate Soundness, % Loss 30.0 25.0 20.0 15.0 10.0 5.0 0.0

Poor

Fair
Performance

Good

Figure 6.3.1-2:

Magnesium Sulfate Soundness vs. Performance 168

Table 6.3.1-2: MSS Success Rate

Trial 1 2 3 4 5 6 7 8 MSS 0% 5% 10% 15% 16% 20% 21% 35% Good 0% 61% 82% 89% 92% 95% 97% 100% Fair 100% 75% 25% 25% 25% 25% 25% 0% Poor 100% 100% 40% 40% 40% 0% 0% 0% Overall 19% 66% 72% 79% 81% 79% 81% 81%

A wealth of research has indicated the L.A. Abrasion test is a poor indicator of field performance. Figure 6.3.1-3 and Table 6.3.1-3 helped to substantiate this

hypothesis. Although the optimized solution (refer to trial 5 of Table 6.3.1-3) is only 1% lower than the optimized solution from micro-Deval, this is mainly due to 80% of the points being good performers and one poor performer having a very high loss value. While the vast majority of poor and fair performers had losses less than 32%, a rather large percentage of good performers had higher losses.
70 60 50 40 30 20 10 0

LA Abrasion, % Loss

Poor

Fair

Good

Perform ance

Figure 6.3.1-3:

L.A. Abrasion vs. Performance

169

Table 6.3.1-3: LAA Success Rate

Trial 1 2 3 4 5 6 LAA 20% 30% 40% 50% 55% 60% Good 31% 64% 92% 95% 100% 100% Fair 50% 25% 0% 0% 0% 0% Poor 83% 33% 17% 17% 17% 0% Overall 39% 57% 76% 78% 82% 80%

Some researchers have suggested that unconfined freeze-thaw testing of aggregate provided no more information than sulfate soundness testing (Boucher and Selig, 1987). A comparison of Figure 6.3.1-2 and Figure 6.3.1-4 seems to indicate Canadian freezethaw testing provided a better indicator of performance than magnesium sulfate soundness testing. While there were no clear lines of division, there appeared to be more of a trend in the freeze-thaw test results. Trial 4 of Table 6.3.1-4 showed the optimized solution yielded an overall success rate of 88%; 7% higher than the total percentage of good performers indicated in trial 7. Although Canadian freeze-thaw appeared to be a better indicator of performance than magnesium sulfate soundness, the results of the test alone were still inconclusive.

170

14.0 12.0 Canadian Freeze-Thaw, % Loss 10.0 8.0 6.0 4.0 2.0 0.0

Poor

Fair
Performance

Good

Figure 6.3.1-4:

Canadian Freeze-Thaw vs. Performance

Table 6.3.1-4: CFT Success Rate

Trial 1 2 3 4 5 6 7 CFT 2.0% 3.0% 4.0% 7.0% 8.0% 9.0% 13.0% Good 44% 62% 82% 100% 100% 100% 100% Fair 100% 75% 25% 25% 25% 25% 0% Poor 100% 100% 60% 40% 40% 20% 0% Overall 54% 67% 75% 88% 88% 85% 81%

Aggregate crushing value test results showed little direct correlation to field performance as shown in Figure 6.3.1-5. As indicated in trial 3 of Table 6.3.1-5, the optimized solution is actually lower than the total percentage of good performers as illustrated in trial 6.

171

50 45 Aggregate Crushing Value, % Loss 40 35 30 25 20 15 10 5 0

Poor Fair Good

Perform ance

Figure 6.3.1-5:

Aggregate Crushing Value vs. Performance Table 6.3.1-5: ACV Success Rate

Trial 1 2 3 4 5 6

ACV 15.0% 20.0% 25.0% 30.0% 35.0% 50.0%

Good 23% 46% 77% 95% 95% 100%

Fair 75% 50% 25% 0% 0% 0%

Poor 100% 50% 33% 17% 0% 0%

Overall 37% 47% 67% 78% 76% 80%

Saturated surface dry aggregate crushing value testing revealed results very similar to oven dry aggregate crushing value testing (refer to Figure 6.3.1-6). As

indicated in trial 3 of Table 6.3.1-6, the optimized solution was actually lower than the total percentage of good performers as illustrated in trial 6.

172

60 Aggregate Crushing Value (SSD), % Loss

50

40

30

20

10

Poor

Fair

Good

Perform ance

Figure 6.3.1-6:

Aggregate Crushing Value (SSD) vs. Performance Table 6.3.1-6: WCV Success Rate

Trial 1 2 3 4 5 6

WCV 15.0% 20.0% 25.0% 30.0% 35.0% 50.0%

Good 23% 49% 74% 92% 95% 100%

Fair 75% 50% 25% 0% 0% 0%

Poor 80% 60% 20% 0% 0% 0%

Overall 33% 50% 65% 75% 77% 81%

Figure 6.3.1-7 and Table 6.3.1-7 illustrates the range of absorption values for each category of performance is quite large. Although the majority of good performers had an absorption of less than 2%, so do the majority of poor performers.

173

5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0

Poor

Fair
Performance

Good

Figure 6.3.1-7:

Absorption vs. Performance

Table 6.3.1-7: ABS Success Rate

Trial 1 2 3 4 5 ABS 1.0% 2.0% 3.5% 4.5% 5.0% Good 66% 92% 92% 100% 100% Fair 75% 75% 0% 0% 0% Poor 80% 20% 20% 20% 0% Overall 68% 83% 77% 83% 81%

Figure 6.3.1-8 and Table 6.3.1-8 indicate no clear correlation between field performance and specific gravity test results. The majority of good and poor performers had specific gravity values above 2.50.

174

2.90 2.80 Specific Gravity (Bulk) 2.70 2.60 2.50 2.40 2.30 2.20

Poor

Fair

Good

Perform ance

Figure 6.3.1-8:

Specific Gravity (Bulk) vs. Performance

Table 6.3.1-8: SG(Bulk) Success Rate

Trial 1 2 3 4 SG(Bulk) 2.70 2.60 2.50 2.20 Good 29% 76% 92% 100% Fair 75% 75% 75% 0% Poor 100% 20% 20% 0% Overall 41% 72% 85% 83%

From Figure 6.3.1-9 and Table 6.3.1-9, no correlations exist between particle shape factor and field performance. The most meaningful optimized solution occurred at 2.00 with an overall success rate of only 42%.

175

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00
Poor Fair Good

Perform ance

Figure 6.3.1-9:

Particle Shape Factor vs. Performance

Table 6.3.1-9: PSF Success Rate

Trial 1 2 3 4 PSF 2.00 2.50 2.90 3.90 Good 33% 74% 87% 100% Fair 75% 0% 0% 0% Poor 80% 20% 20% 0% Overall 42% 63% 73% 81%

Table 6.3.1-10 indicates the majority of the poor performers (67%) had a percentage of fractured particles less than 100%. This is somewhat expected as the majority of the failures were due to pop-outs. The high percentage of 100% fractured particles for good and fair is indicative of the fact that most sources were not detrital.

176

Table 6.3.1-10:
% of total 100 90 - 99 80 - 89 70 - 79 60 - 69 50 - 59 40 - 49 30 - 39 20 - 29 10 - 19 0 - 10 POOR overall number % 2 33% 0 0% 1 17% 2 33% 0 0% 0 0% 0 0% 0 0% 1 17% 0 0% 0 0%

Fractured Particles (2 or more sides) vs. Performance

Fractured Particles (2 or more sides) FAIR GOOD Overall % overall overall POOR FAIR GOOD number % number % 3 75% 29 74% 6% 9% 85% 0 0% 2 5% 0% 0% 100% 0 0% 0 0% 100% 0% 0% 0 0% 2 5% 50% 0% 50% 0 0% 2 5% 0% 0% 100% 0 0% 0 0% 0% 0% 0% 0 0% 0 0% 0% 0% 0% 0 0% 1 3% 0% 0% 100% 0 0% 0 0% 100% 0% 0% 0 0% 0 0% 0% 0% 0% 1 25% 3 8% 0% 25% 75%

6.3.2

Combinations Involving Micro-Deval The combination of the following tests with micro-Deval yielded fairly good

indicators of aggregate field performance in portland cement concrete: sulfate soundness, Canadian freeze-thaw, absorption, and specific gravity. Figure 6.3.2-1 and Table 6.3.2-1 illustrates the correlation between the combination of magnesium sulfate soundness and micro-Deval test results and field performance. As indicated by the solid lines, the limits of a magnesium sulfate soundness loss of 5% and microDeval loss of 9% provided minimum values for which all aggregates with corresponding values below both limits were good performers.

magnesium

Table 6.3.2-1: MSS vs. MD Success Rate

Trial MSS MD Quadrant Good Fair Poor Overall 1 5.0% 9% 100% 45% 100% 100% 55% 2 16.0% 21% 90% 92% 50% 60% 85%

However, this quadrant only contained 45% of the good

performers (refer to trial 1 of Table 6.3.2-1). Trial 2 and the dashed lines indicate the overall optimized solution which was bounded by loss limits of 16% and 21% respectively. This solution successfully qualified 85% of the aggregates; 2% higher than

177

micro-Deval alone. It also much better qualified the poor performers; increasing from 20% to 60%.
35.0 Magnesium Sulfate Soundness, % Loss 30.0 25.0 20.0 15.0 10.0 5.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.3.2-1:

Magnesium Sulfate Soundness vs. Micro-Deval

The combination of L.A. Abrasion and micro-Deval did not yield good correlations with field performance in portland cement concrete. As illustrated in Figure 6.3.2-1, the points were well scattered with little grouping by performance.

178

70 60 LA Abrasion, % Loss 50 40 30 20

POOR
10 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

FAIR GOOD
40.0

Figure 6.3.2-2:

L.A. Abrasion vs. Micro-Deval

Figure 6.3.2-3 and Table 6.3.2-2 illustrate the substantiation of Senior and Rogers previous findings on a correlation between field performance and micro-Deval and Canadian freeze-thaw test results (Senior and Rogers, 1991). As detailed by the solid lines and trial 1, the

Table 6.3.2-2: CFT vs. MD Success Rate

Trial CFT MD Quadrant Good Fair Poor Overall 1 3.6% 21% 100% 77% 100% 100% 81% 2 6.5% 21% 90% 95% 50% 60% 88%

majority of the good performers were bounded by a Canadian freeze-thaw loss of less than 3.6% and a micro-

Deval loss of less than 21% with no fair or poor performers. These bounds successfully qualified 81% of the total. Increasing the limit of the Canadian freeze-thaw loss to 6.5% (refer to dashed lines and trial 2) decreased the accuracy within the bounds to 90% and the accuracy of qualifying fair and poor performers to 56%, but increased the qualification of good performers to 95% and the overall qualification to 88%. This was a 179

5% improvement over the overall success rate for micro-Deval alone and was due to a 40% improvement in the qualification success rate of poor performers.

14.0 12.0 Canadian Freeze-Thaw, % Loss 10.0 8.0 6.0 4.0 2.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.3.2-3:

Canadian Freeze-Thaw vs. Micro-Deval

As was shown in Figure 6.3.2-4, the combination of aggregate crushing value and micro-Deval did not yield good correlations with field performance in portland cement concrete.

180

50 45 Aggregate Crushing Value, % Loss 40 35 30 25 20 15 10 5 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure 6.3.2-4:

Aggregate Crushing Value vs. Micro-Deval

Figure 6.3.2-5 displays very similar results between aggregate crushing value and aggregate crushing value (SSD). This is to be expected as loss values did not change drastically with most aggregates whether the conditions are oven dry or saturated surface dry.

181

60 Aggregate Crushing Value (SSD), % Loss 50 40

30 20

POOR
10 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

FAIR GOOD

Figure 6.3.2-5:

Aggregate Crushing Value (SSD) vs. Micro-Deval Table 6.3.2-3: ABS vs. MD Success Rate
Trial ABS MD Quadrant Good Fair Poor Overall 1 1.2% 18% 96% 58% 75% 100% 64% 2 1.2% 21% 93% 68% 75% 80% 70% 3 2.0% 18% 91% 79% 75% 60% 77% 4 2.0% 21% 89% 89% 75% 40% 83%

As Figure 6.3.2-6 illustrates, good correlations were observed

between performance and absorption and micro-Deval loss. Having a

micro-Deval loss of less than 18% and absorption of less than 1.2%, as

displayed by the solid lines and trial 1 of Table 6.3.2-3, usually indicated a good performing aggregate. These bounds successfully qualified 96% of the aggregates within the limits, but only 59% of the good performers and 65% of the total. Increasing the absorption limit to 2% and the micro-Deval limit to 21% as detailed by the dashed lines and trial 4, increased the total successful qualifying rate to 83% and the qualification of good performers to 90%, but decreased the rate within the quadrant to 90%. Compared 182

to micro-Deval alone, the overall rate did not improve, and the qualification rate for good performers actually declined (95% to 90%), but the identification of fair and poor performers increased from 50% and 20% respectively to 75% and 40%.
5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure 6.3.2-6:

Absorption vs. Micro-Deval

Figure 6.3.2-7 and Table 6.3.2-4 indicate a fairly good correlation with field performance when microDeval results are combined with specific gravity (bulk). The overall success rate was better than the rate for absorption and micro-Deval. This was due to the

Table 6.3.2-4: SG(Bulk) vs. MD Success Rate

Trial SG(Bulk) MD Quadrant Good Fair Poor Overall 1 2.45 21% 90% 95% 75% 40% 87%

increased success rate for identifying good performers; from 90% to 92%.

183

3.00 2.90 2.80 Specific Gravity (Bulk) 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure 6.3.2-7:

Specific Gravity (Bulk) vs. Micro-Deval

The combination of particle shape factor and micro-Deval did not yield good correlations with field performance in portland cement concrete. As illustrated in Figure 6.3.2-8, the points were well scattered with little grouping by performance.

184

4.00 3.50 Particle Shape Factor

3.00 2.50 2.00

1.50 1.00 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure 6.3.2-8:

Particle Shape Factor vs. Micro-Deval

There were no apparent correlations between micro-Deval and percentage of fractured particles tests as illustrated in Figure 6.3.2-9.

185

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure 6.3.2-9: 6.3.3

Fractured Particles (2 or more sides) vs. Micro-Deval

Other Relevant Combinations Figure 6.3.3-1 and Table 6.3.3-1 Table 6.3.3-1: ABS vs. CFT Success Rate
Trial ABS CFT Quadrant Good Fair Poor 1 2 2.0% 3.4% 100% 64% 100% 100% 3 2.0% 6.3% 90% 92% 75% 40%

indicated a good correlation between field performance and Canadian freeze-thaw and absorption results. The delineation indicated by the solid lines and trial 1 accurately identified 80% of the total aggregates. Limits could also

97% 78% 75% 100%

have been drawn with straight lines at a Canadian freeze-thaw loss of less than 3.4% and an absorption of less than 2% (refer to the dashed lines and trial 2) which would have included only good performers, but this would only have accurately qualified 71% of the total. A final delineation could have been drawn at a Canadian freeze-thaw loss of 6.3% and absorption 2% (refer to double-dashed lines and Trial 3) which had an overall success rate of 85%. 186

5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

Figure 6.3.3-1: 6.3.4 Results Summary

Absorption vs. Canadian Freeze-Thaw

From Table 6.3.4-1, combinations of tests did not improve the overall success rate considerably for predicting field performance in portland cement concrete. However, the success rates for the quadrants were quite good when micro-Deval was combined with magnesium sulfate soundness, Canadian freeze-thaw, absorption, or specific gravity. Micro-Deval and Canadian freeze-thaw represented the best combination; resulting in 77% of the good performers being isolated from all fair and poor performers. Table 6.3.4-1: PCC Success Rate Summary
Test Micro-Deval Magnesium Sulfate Soundness Canadian Freeze-Thaw Absorption Specific Gravity (Bulk) Test Alone 83% 81% 88% 83% 85% Micro-Deval Combination N/A 85% 88% 83% 87%

187

6.3.5

Limestone and Dolomite

6.3.5.1 Combinations Involving Micro-Deval From Figure 6.3.5.1-1 and Table 6.3.5.1-1, the correlations for this sub-group and micro-Deval alone were not as good as the whole. From trial 5, the micro-Deval limit of 21% still held true as the optimum limit, but the overall success rate dropped from 83% to 75%.
40.0 35.0 30.0 Micro-Deval, % Loss 25.0 20.0 15.0 10.0 5.0 0.0

Poor

Fair

Good

Perform ance

Figure 6.3.5.1-1:

Micro-Deval vs. Performance MD Success Rate

Fair 100% 67% 67% 67% 67% 33% 0% 0% Poor 100% 100% 100% 0% 0% 0% 0% 0% Overall 33% 42% 50% 58% 75% 67% 58% 67%

Table 6.3.5.1-1:
Trial 1 2 3 4 5 6 7 8 MD 5% 10% 15% 20% 21% 25% 30% 35% Good 0% 25% 38% 63% 88% 88% 88% 100%

188

The combination of magnesium sulfate soundness and micro-Deval provided results for limestone and dolomite similar to the comprehensive results. From

Table 6.3.5.1-2: MSS vs. MD Success Rate

Trial MSS MD Quadrant Good Fair Poor Overall 1 5.0% 9% 100% 13% 100% 100% 42% 2 16.0% 21% 86% 75% 67% 100% 75%

Figure 6.3.5.1-2 and Table 6.3.5.1-2, the overall percentages for trial 1 and 2 dropped from 55% and 85% to 42% and 75%, respectively. This was due to the good performers only making up 67% of the limestone and dolomites versus 81% of all of the aggregates.
30.0 Magnesium Sulfate Soundness, % Loss

25.0

20.0

15.0

10.0

5.0

POOR FAIR GOOD

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0

0.0 Micro-Deval, % Loss

Figure 6.3.5.1-2:

Magnesium Sulfate Soundness vs. Micro-Deval

189

As was shown in Figure 6.3.5.1-3 and Table 6.3.5.1-3, the bounds of a micro-Deval loss less than 21% and a Canadian freeze-thaw loss of less than 6.5% (refer to solid lines and trial 2), as were indicated in the comprehensive section, qualified 88% of the good performers for limestone and dolomite. The correlations previously discussed for the comprehensive results held true for limestone and dolomite, as well.
14.0 12.0 Canadian Freeze-Thaw, % Loss 10.0 8.0 6.0 4.0 2.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0

Table 6.3.5.1-3: CFT vs. MD Success Rate

Trial CFT MD Quadrant Good Fair Poor Overall 1 3.6% 21% 100% 38% 100% 100% 58% 2 6.5% 21% 88% 88% 67% 100% 83%

POOR FAIR GOOD

30.0

35.0

40.0

Micro-Deval, % Loss

Figure 6.3.5.1-3:

Canadian Freeze-Thaw vs. Micro-Deval

190

Absorption

versus

microFrom

Table 6.3.5.1-4:
Trial ABS MD Quadrant Good Fair Poor Overall

Deval did not yield much.

ABS vs. MD Success Rate

2 1.2% 21% 60% 38% 67% 0% 42% 3 2.0% 18% 80% 50% 67% 0% 50% 4 2.0% 21% 71% 63% 67% 0% 58%

Figure 6.3.5.1-4 and Table 6.3.5.1-4, the limits discussed section, in the

comprehensive

absorption

less than 2% and micro-Deval loss less than 21%, only successfully

1 1.2% 18% 67% 25% 67% 0% 33%

qualified 71% within the quadrant and 58% of the total for limestone and dolomite. The best overall success rate was achieved at a limit of 11% loss for micro-Deval and was independent of absorption (refer to blue line).

5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.3.5.1-4:

Absorption vs. Micro-Deval

191

The results for specific gravity versus microDeval for limestone and dolomite were similar to the comprehensive results. From Figure 6.3.5.1-5 and Table 6.3.5.1-5, the quadrant successfully qualified 88% of the good performers and 75% overall, but it only correctly qualified 50% of the fair and poor performers combined. Note this is independent of specific gravity as the microDeval loss limit of 11% defined the same boundary.

Table 6.3.5.1-5: SG(Bulk) vs. MD Success Rate

Trial SG(Bulk) MD Quadrant Good Fair Poor Overall 1 2.45 21% 78% 88% 67% 0% 75%

3.00 2.90 2.80 Specific Gravity (Bulk) 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.3.5.1-5:

Specific Gravity (Bulk) vs. Micro-Deval

192

6.3.5.2 Other Relevant Combinations The correlations provided in the comprehensive section between field performance and absorption and Canadian freeze-thaw test results were not very accurate for limestone and dolomite. From Figure 6.3.5.2-1 and Table 6.3.5.2-1, the overall success rates for trials 1 and 2 dropped 30% and 21% respectively. This was due to the lack of good performers with low Canadian freeze-thaw losses. Trial 3 remained fairly accurate despite 37% of the good performers falling outside the quadrant. As illustrated by the blue lines and trial 4, the overall success was improved to 83% by increasing the absorption limit to 4.5%.
5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Canadian Freeze-Thaw, % Loss

Table 6.3.5.2-1:
Trial ABS CFT Quadrant Good Fair Poor Overall 1

ABS vs. CFT Success Rate

2 2.0% 3.4% 100% 25% 100% 100% 50% 3 2.0% 6.3% 83% 63% 67% 100% 67% 4 2.0% 6.3% 80% 100% 33% 100% 83%

75% 38% 67% 100% 50%

POOR FAIR GOOD

Figure 6.3.5.2-1:

Absorption vs. Canadian Freeze-Thaw 193

6.3.5.3 Section Results Summary The results for this rock type showed it was more difficult to predict when compared to the comprehensive section. From Table 6.3.5.3-1, the success rate could not be improved for any combination of tests over one test alone. This was not entirely unexpected as the material properties of limestones and dolomites can vary quite significantly. Table 6.3.5.3-1:
Test Micro-Deval Magnesium Sulfate Soundness Canadian Freeze-Thaw Absorption Specific Gravity (Bulk)

PCC Limestone and Dolomite Success Rate Summary

Test Alone 75% 75% 83% 58% 67% Micro-Deval Combination N/A 75% 83% 75% 75%

6.3.6

Siliceous Gravel

6.3.6.1 Combinations Involving Micro-Deval As Figure 6.3.6.1-1 and Table 6.3.6.1-1 illustrate, the correlations for the siliceous gravel sub-group were very good. At a limit of 11% (refer to trial 3), the qualification rate for good and poor performers was 100% and the overall success rate was 89%.

194

25.0

20.0 Micro-Deval, % Loss

15.0

10.0

5.0

0.0

Poor

Fair
Performance

Good

Figure 6.3.6.1-1:

Micro-Deval vs. Performance

Table 6.3.6.1-1:
Trial 1 2 3 4 5 6 MD 5% 10% 11% 15% 20% 25% Good 33% 83% 100% 100% 100% 100%

MD Success Rate
Fair 0% 0% 0% 0% 0% 0% Poor 100% 100% 100% 50% 50% 0% Overall 44% 78% 89% 78% 78% 67%

The results of magnesium sulfate and micro-Deval did not show very good correlations with field performance for siliceous gravel. From Table 6.3.6.1-2, the limits of 16% magnesium sulfate loss and 21% microDeval loss from the comprehensive section did provide a small improvement over the total percentage of good 195

Table 6.3.6.1-2: MSS vs. MD Success Rate

Trial MSS MD Quadrant Good Fair Poor Overall 1 5.0% 9% 100% 50% 100% 100% 67% 2 16.0% 21% 75% 100% 0% 50% 78%

performers. However, the best overall success rate (89%) was achieved with a microDeval loss limit of 11% independent of magnesium sulfate loss (refer to blue line).
18.0 Magnesium Sulfate Soundness, % Loss 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.3.6.1-2:

Magnesium Sulfate Soundness vs. Micro-Deval

As was shown in Figure 6.3.6.1-3 and Table

Table 6.3.6.1-3: CFT vs. MD Success Rate

1 3.6% 21% 100% 100% 100% 100% 100% 2 6.5% 21% 75% 100% 0% 50% 78%

6.3.6.1-3, all of the good performers in the siliceous Trial gravel sub-group fit within the overall bounds of 3.6% CFT Canadian freeze-thaw loss and 21% micro-Deval loss as defined in trial 1 for the comprehensive section. These bounds successfully qualified 100% of all categories.
MD Quadrant Good Fair Poor Overall

196

7.0 6.0 Canadian Freeze-Thaw, % Loss 5.0 4.0 3.0 2.0 1.0 0.0 0.0 5.0 10.0 15.0 20.0 Micro-Deval, % Loss

POOR FAIR GOOD

25.0

Figure 6.3.6.1-3:

Canadian Freeze-Thaw vs. Micro-Deval

Figure 6.3.6.1-4 and Table 6.3.6.1-4 showed the overall values indicated

Table 6.3.6.1-4:
Trial ABS MD Quadrant Good Fair Poor Overall 1 1.2% 18% 100% 33% 100% 100% 56%

ABS vs. MD Success Rate

2 1.2% 21% 100% 33% 100% 100% 56% 3 2.0% 18% 86% 100% 100% 50% 89% 4 2.0% 21% 86% 100% 100% 50% 89% 5 2.0% 11% 100% 100% 100% 100% 100%

earlier of a micro-Deval loss less than 21% and an

absorption of less than 2%

indicating a good performer mostly held true for the siliceous gravel sub-group, but could be improved. Within this sub-group, a micro-Deval loss less than 11.0% and an

absorption of less than 2% isolated all of the good performers (refer to blue lines and trial 5). These bounds successfully qualified 100% of all categories.

197

3.5 3.0 2.5 Absorption, % 2.0 1.5 1.0 0.5 0.0 0.0 5.0 10.0 15.0 20.0 25.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.3.6.1-4:

Absorption vs. Micro-Deval

Specific gravity also had very good correlation with field performance when combined with microDeval. From Figure 6.3.6.1-5 and Table 6.3.6.1-5, the Trial limits of specific gravity greater than 2.5 and microDeval loss less than 11%, successfully qualified all of the aggregates (refer to blue lines and trial 2).

Table 6.3.6.1-5: SG(Bulk) vs. MD Success Rate

SG(Bulk) MD Quadrant Good Fair Poor Overall 1 2.45 21% 90% 95% 75% 40% 87% 2 2.50 11% 100% 100% 100% 100% 100%

198

3.00 2.90 2.80 Specific Gravity (Bulk) 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 Micro-Deval, % Loss

POOR FAIR GOOD

25.0

Figure 6.3.6.1-5:

Specific Gravity (Bulk) vs. Micro-Deval

6.3.6.2 Other Relevant Combinations Figure 6.3.6.2-1 and Table 6.3.6.2-1 indicated a correlation Table 6.3.6.2-1:
Trial ABS CFT Quadrant Good Fair Poor Overall 1

ABS vs. CFT Success Rate

2 2.0% 3.4% 100% 64% 100% 100% 71% 3 2.0% 6.3% 90% 92% 75% 40% 85% 4 2.0% 4.0% 86% 100% 100% 50% 89%

similar to the one described for the comprehensive section. The solid

line and trial 1 delineated 100% of the points below the line, but only 78% overall. The best overall

97% 78% 75% 100% 80%

success rate was achieved with an absorption less than 2% and a Canadian freeze-thaw loss less than 4% (refer to blue lines and trial 4). These limits produced an overall success rate of 89%.

199

4.0 3.5 3.0 Absorption, % 2.5 2.0 1.5 1.0 0.5 0.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

Figure 6.3.6.2-1: 6.3.6.3 Section Results Summary

Absorption vs. Canadian Freeze-Thaw

The results for the siliceous gravel sub-group showed combinations of Canadian freeze-thaw, absorption, or specific gravity with micro-Deval were strong predictors of field performance in portland cement concrete. As summarized in Table 6.3.6.3-1, all three of these combinations resulted in perfect performance prediction. Table 6.3.6.3-1:
Test Micro-Deval Magnesium Sulfate Soundness Canadian Freeze-Thaw Absorption Specific Gravity (Bulk)

PCC Siliceous Gravel Success Rate Summary

Test Alone 89% 67% 89% 78% 78% Micro-Deval Combination N/A 89% 100% 100% 100%

200

6.4

Performance Analysis for Base Course Aggregates

6.4.1 Individual Tests A considerable number of aggregates studied in this research project were also used in base courses throughout the United States and Canada. Hence, attempts will be made to determine the prediction success rate of the tests performed in this research study for aggregates used in base courses. However, it should be noted that the vast majority of the aggregates are good performers. Only two fair performers and three poor performers were identified by participating DOTs. Therefore, the analysis procedure used for hotmix asphalt aggregates cannot be directly applied in this situation. The overall success rate cannot be considered the determining factor in evaluating the prediction success rate of a test alone since the highest percentage will be obtained at a threshold value such that all good sources fall below it regardless of where fair and poor sources lie. However, the quadrant quantitative analysis previously performed does provide meaningful information for two-test combinations and will be used. Base course performance graphs for each test are provided in Appendix O if the reader should choose to study them. Thus, the following sections will provide qualitative and quantitative analysis for two-test combinations and attempt to draw conclusions as to which two-test combination provides the highest prediction success rate.

6.4.2

Combinations Involving Micro-Deval Two-test combinations were studied in this research project by plotting the results

of every test against every other test. The resulting graphs are provided in Appendix O. In this section, only the plots involving micro-Deval will be quantitatively as well as qualitatively analyzed since the main purpose of this research project is to evaluate the 201

effectiveness of predictions using micro-Deval. As previously discussed, due to only two fair and three poor performers present, close attention must now be paid to the quadrant success rate with secondary importance placed on computed overall success rate. Nevertheless, conclusions can be drawn on the extent of successful aggregate qualification produced by micro-Deval in combination with another test versus microDeval alone. Figure 6.4.2-1 demonstrates the results of micro-Deval plotted versus magnesium sulfate soundness results. Additionally, Table 6.4.2-1 summarizes the
80.0 Magnesium Sulfate Soundness, % Loss 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.4.2-1:

Magnesium Sulfate Soundness vs. Micro-Deval

results of the performed quantitative analysis using Microsoft Excel 2003. As the data illustrate, the overall success rate can be significantly increased by expanding the quadrant laterally to the right. Trial #3 represents the highest overall success rate

achieved through extensive trial and error process. More importantly, however, the

202

Table 6.4.2-1: MSS and MD Success Rate

TRIAL NUMBER MSS MD QUADRANT GOOD FAIR POOR OVERALL 1 20 12 100% 44% 100% 100% 48% 2 21 19.5 96% 78% 50% 67% 77% 3 21 21 94% 83% 0% 67% 80%

quadrant success rate only falls 6% throughout the process, with 94% success rate in the optimal situation. However, although 83% of good performers are included at this point, none of the fair performers are identified correctly. Overall, it can be concluded that micro-Deval in addition to magnesium sulfate provide good prediction results. Figure 6.4.2-2 demonstrates the results of micro-Deval losses plotted against L.A. Abrasion losses. Three quadrants were attempted to optimize the overall performance as
70

POOR
60 LA Abrasion, % Loss 50

FAIR GOOD

40 30

20 10 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

Figure 6.4.2-2:

L.A. Abrasion vs. Micro-Deval 203

well as show how quadrant success rate changes as the quadrant is expanded to the right. The results are summarized in Table 6.4.2-2. As the data illustrate, a maximum overall success rate of 80% can be achieved using micro-Deval loss of 22% and L.A. Abrasion loss of 65%. The reader should note that the quadrant success rate, which is more

Table 6.4.2-2: LAA and MD Success Rate

TRIAL NUMBER LAA MD QUADRANT GOOD FAIR POOR OVERALL 1 65 12 100% 44% 100% 100% 48% 2 65 13.5 97% 54% 100% 67% 56% 3 65 22 93% 85% 0% 33% 80%

important in our analysis due to very few poor and fair sources present, decreases as overall success rate increases. However, the decrease for each consecutive trial are very low, with the maximum difference of 7% between Trial #1 and Trial #3, which represents the maximum overall success rate. Thus, the final quadrant success rate is 93% and, although relatively high, is less than the 94% achieved through micro-Deval and magnesium sulfate soundness combination. Additionally, although 85% of good sources are fall within the quadrant, none of the fair sources are identified correctly. Figure 6.4.2-3 demonstrates the results of Canadian freeze-thaw test plotted against micro-Deval losses. Quantitative analysis is performed using Microsoft Excel,

204

25.0

Canadian Freeze-Thaw, % Loss

20.0

15.0

10.0

5.0

POOR FAIR GOOD

0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

Figure 6.4.2-3:

Canadian Freeze-Thaw vs. Micro-Deval Since both magnesium sulfate

and the results are summarized in Table 6.4.2-3.

soundness and Canadian freeze-thaw tests measure the aggregate resistance to freezethaw cycles, similar results were expected. variability observed Moreover, due to extremely high test

Table 6.4.2-3: CFT and MD Success Rate

TRIAL NUMBER CFT MD QUADRANT GOOD FAIR POOR OVERALL 1 10 12 100% 44% 100% 100% 48% 2 12 13 97% 53% 100% 67% 55% 3 12.5 21 96% 83% 50% 67% 81%

205

during magnesium sulfate testing, greater success rates were expected. As Table 6.4.2-3 indicates, the expectations are substantiated by the computed data. The maximum overall success rate of 81% is achieved as shown in Trial #3 but at the expense of decreasing quadrant success rate. However, to achieve the optimal overall success rate, the quadrant success rate is only reduced by 4% to 96%, which is better than 94% observed through utilization of micro-Deval and magnesium sulfate soundness data. Thus, this

combination produces the highest overall success rate thus far of 81% while also achieving the highest quadrant success rate of 96%. Figure 6.4.2-4 and Table 6.4.2-4 summarize the results of aggregate crushing value losses plotted against micro-Deval losses. maximum overall
45 Aggregate Crushing Value, % Loss 40 35 30 25 20 15 10 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

The results demonstrate that the

POOR FAIR GOOD

Figure 6.4.2-4:

Aggregate Crushing Value vs. Micro-Deval

success rate of 80% is achieved through this two-test combination as shown by Trail #3. Furthermore, quadrant success rate of 94% is achieved in the same trial as well, 206

indicating that optimal results are achieved at the expense of only 6% decrease in the quadrant success rate. Although the success rates are not as high as those achieved through Canadian freeze-thaw and micro-Deval combination, they are still quite high and indicate that useful information is gained through the usage of this two-test combination. Table 6.4.2-4: ACV and MD Success Rate
TRIAL NUMBER ACV MD QUADRANT GOOD FAIR POOR OVERALL 1 33 12 100% 44% 100% 100% 48% 2 33 13.5 97% 54% 100% 67% 56% 3 33 21 94% 83% 0% 67% 80%

The plot of saturated, surface-dry aggregate crushing value test versus microDeval is shown in Figure 6.4.2-5.
45 Aggregate Crushing Value (SSD), % Loss 40 35 30 25 20 15 10 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.4.2-5:

Aggregate Crushing Value (SSD) vs. Micro-Deval 207

Since the test is very similar to the aggregate crushing value test previously analyzed, similar results were expected. Table 6.4.2-5 summarizes the results of the quantitative analysis carried out on the plot data. Both the quadrant success rate of 94% and overall

Table 6.4.2-5: WCV and MD Success Rate

TRIAL NUMBER WCV MD QUADRANT GOOD FAIR POOR OVERALL 1 33 12 100% 44% 100% 100% 48% 2 33 19.5 96% 78% 50% 67% 77% 3 33 21 94% 83% 0% 67% 80%

success rate of 80% are identical to that of aggregate crushing value and micro-Deval combination despite the different trial course of action taken. Thus, it can be concluded that this combination does not provide any more information than the previouslydiscussed aggregate crushing value and micro-Deval combination. Figure 6.4.2-6 provides the results of incorporation micro-Deval and absorption information on the same plot. Quantitative analysis carried out using Microsoft Excel 2003 is summarized in Table 6.4.2-6. At micro-Deval loss of 22% and absorption capacity of 6%, the overall success rate of 80% is achieved while the quadrant success rate is 93%. However, at Trial #3 which represents the optimal solution for the overall success rate as well as the quadrant success rate, none of the fair aggregates are identified correctly and only one poor source is identified as poor. Thus, it can be concluded that the combination does provide useful information but it is not as high as achieved by other tests.

208

6.0

5.0

4.0 Absorption

3.0

2.0

POOR
1.0

FAIR GOOD

0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

Figure 6.4.2-6:

Absorption vs. Micro-Deval

Table 6.4.2-6: ABS and MD Success Rate

TRIAL NUMBER ABS MD QUADRANT GOOD FAIR POOR OVERALL 1 2 12 100% 44% 100% 100% 48% 2 5 19 96% 75% 50% 67% 73% 3 6 22 93% 85% 0% 33% 80%

Bulk specific gravity, saturated surface dry specific gravity, as well as apparent specific gravity were computed during this research project, and the graphs for all three are included in Appendix O. However, all three exhibit identical results and only bulk specific gravity versus micro-Deval is shown in Figure 6.4.2-7 as well as further analyzed with Table 6.4.2-7 providing a summery of the quantitative analysis. As the table 209

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure 6.4.2-7:

Specific Gravity (Bulk) vs. Micro-Deval

indicates, the highest overall success rate achieved is once again 80%, and the highest quadrant success rate is 93%. Once again, a very high success rate of 85% of good sources is achieved in the optimal solution while none of the fair sources

Table 6.4.2-7: SG (Bulk) and MD Success Rate

TRIAL NUMBER SG(BULK) MD QUADRANT GOOD FAIR POOR OVERALL 1 3.05 12 100% 44% 100% 100% 48% 2 3.05 19.5 96% 78% 50% 67% 77% 3 3.05 21 94% 83% 0% 67% 80%

210

are identified correctly and only one poor sources is qualified as such. Thus, it can be concluded that this two-test combination provides useful information although not as much as previously-discussed micro-Deval and Canadian freeze-thaw tests combination. The results of particle shape factor and micro-Deval are shown in Figure 6.4.2-8 and summarized in Table 6.4.2-8. As the data demonstrate, the same overall
4.50

POOR
4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

FAIR GOOD

Particle Shape Factor

Figure 6.4.2-8:

Particle Shape Factor vs. Micro-Deval

success rate of 80% was achieved. However, this two-test combination also provides the highest quadrant success rate observed thus far of 94%. Although none of the fair sources are identified correctly, 83% of good sources are and more importantly two of the three poor sources are identified correctly as well. Thus, it can be concluded that the information provided by this combination is quite significant.

211

Table 6.4.2-8: PSF and MD Success Rate

TRIAL NUMBER PSF MD QUADRANT GOOD FAIR POOR OVERALL 1 4 12 100% 44% 100% 100% 48% 2 4 19.5 96% 78% 50% 67% 77% 3 4 21 94% 83% 0% 67% 80%

6.4.3

Other Relevant Combinations Since every test was plotted against every other test carried out during this

research project, it was necessary to visually inspect all graphs and perform qualitative analysis on those graphs that visual inspection identified as possibly having a high quadrant success rate. Only the following combinations exhibited the possibility of meaningful quadrant division: Aggregate crushing value and L.A. abrasion Particle shape factor and L.A. abrasion Particle shape factor and Canadian freeze-thaw

Quantitative analysis was performed on those three graphs according to the procedure previously described. However, in each situation, it was determined that the optimal solution lies when all points lie within the quadrant. Hence, no useful

information for comparison purposes can be extracted for the three plots.

212

6.4.4

Results Summary Table 6.4.4-1 provides a summary of the discussion above. As the data illustrate,

all two-test combinations had very similar overall success rates with the maximum of 81% being achieved by micro-Deval and Canadian freeze-thaw combination. Table 6.4.4-1: BC Success Rates Summary
OVERALL SUCCESS RATE 80% 80% 81% 80% 80% 80% 80% 80% QUADRANT SUCCESS RATE 94% 93% 96% 94% 94% 93% 93% 94%

MAGNESIUM SULFATE SOUNDNESS L.A. ABRASION CANADIAN FREEZE-THAW AGGREGATE CRUSHING VALUE AGGREGATE CRUSHING VALUE (SSD) ABSORPTION SPECIFIC GRAVITY (BULK) PARTICLE SHAPE FACTOR

This test combination also represents the highest quadrant success rate of 96%. All other tests had the same overall success rate of 80%. Four two-test combinations had a quadrant success rate of 94% while three had a quadrant success rate of 93%. Although the data illustrate that better information can be gained by certain two-test combinations, more research is necessary where fair and poor sources are represented more evenly.

213

6.5 6.5.1

Performance Analysis for Open-Graded Friction Course Aggregates Individual Tests From Figure 6.5.1-1 and Table 6.5.1-1, the majority of micro-Deval loss values

for open-graded friction course fall below 15%. As indicated by trial 3, 15% loss was the optimized solution which represented an increase of 5% over the total percentage of good performers. While not conclusive, an 18% loss limit as specified by Kandhal and Parker (Kandhal and Parker, 1998) appeared to separate the majority of the good performers from the majority of the fair and poor performers.
30.0

25.0

Micro-Deval, % Loss

20.0

15.0

10.0

5.0

0.0

Poor

Fair
Performance

Good

Figure 6.5.1-1:

Micro-Deval vs. Performance

214

Table 6.5.1-1: MD Success Rate

Trial 1 2 3 4 5 6 MD 5% 10% 15% 20% 25% 30% Good 12% 59% 88% 88% 94% 100% Fair 100% 100% 50% 0% 0% 0% Poor 100% 100% 67% 67% 0% 0% Overall 32% 68% 82% 77% 73% 77%

The results for magnesium sulfate soundness alone were similar to micro-Deval with an optimized success rate of 82%. From Figure 6.5.1-2 and Table 6.5.1-2, the optimized success rate was achieved at trial 1, 2, and 5 with magnesium sulfate soundness losses of 5%, 10%, and 30%, respectively. The fair performer with an

extremely high loss, denoted source #2, was sandstone which did not hold up well to the magnesium sulfate solution.
80.0 Magnesium Sulfate Soundness, % Loss 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0

Poor

Fair
Performance

Good

Figure 6.5.1-2:

Magnesium Sulfate Soundness vs. Performance

215

Table 6.5.1-2: MSS Success Rate

Trial 1 2 3 4 5 6 MSS 5% 10% 15% 20% 30% 70% Good 82% 88% 88% 94% 100% 100% Fair 50% 50% 50% 50% 50% 0% Poor 100% 67% 0% 0% 0% 0% Overall 82% 82% 73% 77% 82% 77%

L.A. Abrasion resulted in the exact same overall success rate as micro-Deval and magnesium sulfate soundness; however, with L.A. Abrasion the high success rate was due solely to source #2. As was shown in Figure 6.5.1-3, source #2, sandstone, had a very high L.A. Abrasion loss which set it well apart from the other sources. Overlooking this solution, the best success rate for L.A. Abrasion was 77% (refer to Table 6.5.1-3, trial 2 and 4); no better than the total percentage of good performers.
70 60 50 40 30 20 10 0

LA Abrasion, % Loss

Poor

Fair
Performance

Good

Figure 6.5.1-3:

L.A. Abrasion vs. Performance

216

Table 6.5.1-3: LAA Success Rate

Trial 1 2 3 4 5 6 LAA 21% 23% 29% 37% 43% 66% Good 53% 76% 82% 94% 100% 100% Fair 100% 100% 50% 50% 50% 0% Poor 100% 67% 33% 0% 0% 0% Overall 64% 77% 73% 77% 82% 77%

As illustrated in Figure 6.5.1-4 and Table 6.5.1-4, the results for Canadian freezethaw alone was not conclusive. The optimized solution represented no improvement in overall success rate from the total percentage of good performers. Similar to the results from the portland cement concrete application, the majority of the good performers fell below 6% loss, but so did a large percentage of fair and poor performers.
10.0 9.0 Canadian Freeze-Thaw, % Loss 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0

Poor

Fair
Performance

Good

Figure 6.5.1-4:

Canadian Freeze-Thaw vs. Performance

217

Table 6.5.1-4: CFT Success Rate

Trial 1 2 3 4 5 CFT 2% 3% 6% 7% 10% Good 35% 65% 94% 94% 100% Fair 100% 50% 0% 0% 0% Poor 67% 33% 33% 0% 0% Overall 45% 59% 77% 73% 77%

Similar to Canadian freeze-thaw, aggregate crushing value had a success rate no better than the percentage of good performers. From Figure 6.5.1-5 and Table 6.5.1-5, the optimized solution occurred at a loss limit of 22% and 24%.
35 30 Aggregate Crushing Value, % Loss 25 20 15 10 5 0

Poor

Fair
Performance

Good

Figure 6.5.1-5:

Aggregate Crushing Value vs. Performance

218

Table 6.5.1-5: ACV Success Rate

Trial 1 2 3 4 5 6 ACV 17% 22% 23% 24% 27% 31% Good 47% 88% 88% 94% 94% 100% Fair 100% 50% 50% 50% 0% 0% Poor 100% 33% 0% 0% 0% 0% Overall 59% 77% 73% 77% 73% 77%

The results for aggregate crushing value (SSD) provided a success rate slightly higher than the total percentage of good performers. As was shown in Figure 6.5.1-6 and Table 6.5.1-6, the optimized solution occurred at a loss of 25% which represented a 2% improvement.
35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0

Poor

Fair
Performance

Good

Figure 6.5.1-6:

Aggregate Crushing Value (SSD) vs. Performance

219

Table 6.5.1-6: WCV Success Rate

Trial 1 2 3 4 5 WCV 18% 23% 25% 30% 32% Good 41% 76% 88% 94% 100% Fair 100% 100% 50% 0% 0% Poor 100% 50% 50% 0% 0% Overall 57% 79% 83% 76% 81%

From Figure 6.5.1-7 and Table 6.5.1-7, the optimized solution for absorption represented a 14% improvement over the total percentage of good performers. As was shown in the figure, only two points, one fair and one poor prevented absorption from correctly categorizing all the sources.
3.0

2.5

2.0 Absorption, %

1.5

1.0

0.5

0.0

Poor

Fair
Performance

Good

Figure 6.5.1-7:

Absorption vs. Performance

Table 6.5.1-7: ABS Success Rate

Trial 1 2 3 ABS 0.5% 2.2% 2.7% Good 24% 100% 100% Fair 100% 50% 0% Poor 100% 67% 0% Overall 41% 91% 77%

220

The results for specific gravity represented fairly good correlations to field performance in open-graded friction course. As indicated in Figure 6.5.1-8 and Table 6.5.1-8, the optimized success rate provided an increase of 9% (refer to trial 3).

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20

Poor

Fair
Performance

Good

Figure 6.5.1-8:

Specific Gravity (Bulk) vs. Performance

Table 6.5.1-8: SG(Bulk) Success Rate

Trial 1 2 3 4 SG(Bulk) 2.70 2.60 2.50 2.20 Good 35% 76% 100% 100% Fair 100% 50% 50% 0% Poor 100% 67% 33% 0% Overall 50% 73% 86% 77%

221

From Figure 6.5.1-9 and Table 6.5.1-9, the results for particle shape factor provided no improvement over the total percentage of good performers.
3.50

3.00 2.50 Particle Shape Factor

2.00

1.50

1.00

0.50 0.00

Poor

Fair
Performance

Good

Figure 6.5.1-9:

Particle Shape Factor vs. Performance

Table 6.5.1-9: PSF Success Rate

Trial 1 2 3 4 PSF 2.00 2.50 3.00 3.50 Good 18% 88% 94% 100% Fair 50% 50% 50% 0% Poor 100% 0% 0% 0% Overall 32% 73% 77% 77%

222

No significant correlations could be drawn from the percentage of fractured particles as indicated in Table 6.5.1-10. Table 6.5.1-10:
POOR overall number % 2 1 0 0 0 0 0 0 0 0 0 67% 33% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Fractured Particles (2 or more sides) vs. Performance

Fractured Particles (2 or more sides) FAIR overall number % 2 0 0 0 0 0 0 0 0 0 0 100% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% GOOD overall number % 13 3 0 1 0 0 0 0 0 0 0 76% 18% 0% 6% 0% 0% 0% 0% 0% 0% 0% Overall % POOR 12% 25% 0% 0% 0% 0% 0% 0% 0% 0% 0% FAIR 12% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% GOOD 76% 75% 0% 100% 0% 0% 0% 0% 0% 0% 0%

% of total 100 90 - 99 80 - 89 70 - 79 60 - 69 50 - 59 40 - 49 30 - 39 20 - 29 10 - 19 0 - 10

6.5.2

Combinations Involving Micro-Deval The results from magnesium sulfate soundness provided no improvement over the

success rate for micro-Deval alone. From Figure 6.5.2-1, the optimized solution was at a micro-Deval loss of 11%, independent of magnesium sulfate soundness loss.

223

80.0 Magnesium Sulfate Soundness, % Loss 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

POOR FAIR GOOD

Figure 6.5.2-1:

Magnesium Sulfate Soundness vs. Micro-Deval

From

Figure

6.5.2-2

and

Table

6.5.2-1,

the

combination of L.A. Abrasion and micro-Deval provided a fairly good predictor of performance in open-graded friction course. The limits detailed in trial 1, successfully isolated 71% of the good performers with no fair or poor performers.

Table 6.5.2-1: LAA vs. MD Success Rate

Trial LAA MD Quadrant Good Fair Poor Overall 1 23% 14% 100% 71% 100% 100% 77%

224

70 60 50 40 30 20 10 0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

POOR FAIR GOOD

LA Abrasion, % Loss

Figure 6.5.2-2:

L.A. Abrasion vs. Micro-Deval

From Figure 6.5.2-3, the combination of Canadian freeze-thaw and micro-Deval did not improve the correlation from micro-Deval alone. The optimized solution for this combination occurred at a micro-Deval loss of 15% independent of the Canadian freezethaw loss.

225

10.0 9.0 Canadian Freeze-Thaw, % Loss 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

POOR FAIR GOOD

30.0

Figure 6.5.2-3:

Canadian Freeze-Thaw vs. Micro-Deval

Combining aggregate crushing value with microDeval provided fairly good limits for the good performers. From Figure 6.5.2-4 and Table 6.5.2-2, 53% of the good performers could be isolated within the bounds of 18% aggregate crushing value and 15% micro-Deval loss.

Table 6.5.2-2: ACV vs. MD Success Rate

Trial ACV MD Quadrant Good Fair Poor Overall 1 19% 15% 100% 53% 100% 100% 64%

226

35 30 25 20 15 10 5 0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

Aggregate Crushing Value, % Loss

POOR FAIR GOOD

30.0

Figure 6.5.2-4:

Aggregate Crushing Value vs. Micro-Deval

The combination of aggregate crushing value (SSD) with micro-Deval provided even better

Table 6.5.2-3: WCV vs. MD Success Rate

Trial WCV MD Quadrant Good Fair Poor Overall 1 23% 15% 100% 71% 100% 100% 77%

correlations.

From Figure 6.5.2-5 and Table 6.5.2-3,

71% of the good performers could be isolated with the bounds described in trial 1. This provided for an overall success rate of 77%.

227

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

POOR FAIR GOOD

Figure 6.5.2-5:

Aggregate Crushing Value (SSD) vs. Micro-Deval

As the absorption test provided the best overall success rate for individual test, it would be expected that the combination with micro-Deval would provide meaningful results as well. From Figure 6.5.2-6 and Table 6.5.2-4, the bounds of trial 2 correctly categorized 88% of the good performers and had a 94% success rate within the quadrant. However, the best overall success rate was absorption of 2.2% independent of micro-Deval loss.

Table 6.5.2-4: ABS vs. MD Success Rate

Trial ABS MD Quadrant Good Fair Poor Overall 1 0.6% 13% 100% 29% 100% 100% 45% 2 2.2% 15% 94% 88% 50% 100% 86%

228

2.5

2.0

Absorption, %

1.5

1.0

0.5

POOR FAIR GOOD

0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

0.0

Figure 6.5.2-6:

Absorption vs. Micro-Deval

While not as good as absorption and micro-Deval, the combination of specific gravity and micro-Deval had nice correlations as well. As can be seen in Figure 6.5.2-7 and Table 6.5.2-5, 82% of the good performers were isolated with a 93% success rate within the quadrant. These bounds resulted in an 82% overall success rate similar to micro-Deval alone, but with better

Table 6.5.2-5: SG(Bulk) vs. MD Success Rate

Trial SG(Bulk) MD Quadrant Good Fair Poor Overall 1 2.55 15% 93% 82% 50% 100% 82%

categorization of the poor performers.

229

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

POOR FAIR GOOD

30.0

Figure 6.5.2-7:

Specific Gravity (Bulk) vs. Micro-Deval

From Figure 6.5.2-8 and Table 6.5.2-6, the results for particle shape factor and micro-Deval were the best at isolating a quadrant of good performers, resulting in correctly categorizing 76% of the good performers with a 100% success rate within the quadrant.

Table 6.5.2-6: PSF vs. MD Success Rate

Trial PSF MD Quadrant Good Fair Poor Overall 1 2.40 15% 100% 76% 100% 100% 82%

230

3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

Particle Shape Factor

POOR FAIR GOOD

30.0

Figure 6.5.2-8:

Particle Shape Factor vs. Micro-Deval

From Figure 6.5.2-9, no meaningful conclusions were drawn from the combination percentage of fractured particles and micro-Deval.

231

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

POOR FAIR GOOD

30.0

Figure 6.5.2-9: 6.5.3

Fractured Particles (2 or more sides) vs. Micro-Deval

Other Relevant Combinations The combination of aggregate crushing value and Table 6.5.3-1: ACV vs. LAA Success Rate
Trial ACV LAA Quadrant Good Fair Poor Overall 1 20% 25% 100% 59% 100% 100% 68%

L.A. Abrasion provided fairly nice correlations with field performance. From Figure 6.5.3-1 and Table 6.5.3-1, the limits described in trial 1 perfectly isolated 59% of the good performers. However, the overall success rate for this combination was only 68%.

232

35 30 25 20 15 10 5 0 0 10 20 30 40 50 60 LA Abrasion, % Loss

Aggregate Crushing Value, % Loss

POOR FAIR GOOD

70

Figure 6.5.3-1:

Aggregate Crushing Value vs. L.A. Abrasion

From Figure 6.5.3-2 and Table 6.5.3-2, the combination of aggregate crushing value (SSD) and Canadian freeze-thaw provided results very similar to the combination of aggregate crushing value (SSD) and microDeval. The exact same percentage of good performers was isolated, and the overall percentage was within 1%.

Table 6.5.3-2: WCV vs. CFT Success Rate

Trial WCV CFT Quadrant Good Fair Poor Overall 1 23% 4.5% 100% 71% 100% 100% 76%

233

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

9.0 10.0

Figure 6.5.3-2:

Aggregate Crushing Value (SSD) vs. Canadian Freeze-Thaw

The results for absorption and aggregate crushing value showed fairly good isolation of the good performers. As was shown in Figure 6.5.3-3 and Table 6.5.3-3, 59% of the good performers were isolated which resulted in an overall success rate of 67%.

Table 6.5.3-3: ABS vs. ACV Success Rate

Trial ABS ACV Quadrant Good Fair Poor Overall 1 2.3% 19% 100% 59% 100% 100% 67%

234

3.0

2.5

Absorption, %

2.0

1.5

1.0

0.5

0.0 0 5 10 15 20 25 30 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

35

Figure 6.5.3-3: 6.5.4 Results Summary

Absorption vs. Aggregate Crushing Value

For the open-graded friction course application several tests and combinations of tests provided good correlations with field performance. The best individual tests were absorption and specific gravity (bulk), having overall success rates of 91% and 86%, respectively. No combination of tests had a better overall success rate than the rate achieved with absorption alone (refer to Table 6.5.4-1), but several combinations did a very good job of isolating large percentages of good performers. The best combinations of tests were L.A. abrasion, aggregate crushing value (SSD), absorption, specific gravity (bulk), and particle shape factor in combination with micro-Deval and also aggregate crushing value (SSD) in combination with Canadian freeze-thaw. Due to limited amount of fair and poor performers, these correlations were dubious at best.

235

Table 6.5.4-1: OGFC Success Rate Summary

Test Micro-Deval Magnesium Sulfate Soundness L.A. Abrasion Canadian Freeze-Thaw Aggregate Crushing Value Aggregate Crushing Value (SSD) Absorption Specific Gravity (Bulk) Particle Shape Factor Test Alone 82% 82% 82% 77% 77% 83% 91% 86% 77% Micro-Deval Combination N/A 82% 82% 82% 82% 83% 91% 86% 82%

6.6

Test Correlations Analysis

6.6.1 Correlations Analysis Overview One of the objectives of this research project was to determine if any correlations could be drawn between any two tests conducted in the study. To accomplish this task, test results of one test were plotted against test results of another test, regardless of performance rating, according to the matrix shown in Table 6.6.1-1, where a shaded

Table 6.6.1-1: Correlations Graphs

SG (BULK) ACV (SSD) SG (APP) SG (SSD)

MSS

ACV

MSS LAA CFT ACV ACV (SSD) ABS SG (BULK) SG (SSD) SG (APP) PSF

236

ABS

LAA

CFT

MD

rectangle indicates that the graph was plotted. Additionally, all graphs are included in the Appendix Q. Once the graphs were created, correlation analysis was performed on each one to measure the strength of the association between numerical variables by performing linear regression analysis. In cases where no linear relationship clearly existed,

logarithmic, polynomial, power, and exponential regression analyses were performed using Microsoft Excel 2003. During the analysis, the coefficient of determination R2, which measures the proportion of variation that is explained by the independent variable X in the regression model, was computed according to the following formula: Error! Objects cannot be created from editing field codes. where SST measures the variation of the Y-values around their mean Y and SSR explains the variation attributable to the relationship between X and Y. Finally, correlation

coefficient can be computed by taking the square root of the R2 value. Since it is common in research practice to report the coefficient of determination value, R2 will be used throughout this report for comparison and analysis purposes. Upon recommendation of Dr. Zhanmin Zhang of the University of Texas at Austin whose expertise lies in the area of statistical analysis within the field of Civil Engineering, R2 values were computed for three cases: 1. The complete data set with no outliers eliminated. 2. The data set with 99.7% of the values retained and outliers eliminated based on the assumption of normal population distribution and using the interval of

3 *

, where is the sample mean and is the sample

standard deviation.

237

3. The data set with 95% of the values retained and outliers eliminated based on the assumption of normal population distribution and using the interval of 2 * , where once again is the sample mean and is the sample

standard deviation. The complete data set analysis is discussed in Section 6.6.2, while Case 2 is described in Section 6.6.3, and Case 3 is discussed in Section 6.6.4.
6.6.2 Complete Data Set Regression Analysis

During this analysis, the complete data set was used for each test with no outliers eliminated to improve R2 values. Only the graphs where significant correlations were found will be discussed below, but the entire graph set is included in Appendix Q. Additionally, Appendix T includes Microsoft Excel tables used to produce the graphs. Table 6.6.2-1 provides a summary of R2 values for each graph plotted in this analysis. The values provided within the table are the highest possible R2 obtained using linear, logarithmic, second-degree polynomial, power, and exponential regression analysis techniques. The reader should refer to Appendix E for the equations of the best-fit curves that resulted in the highest possible R2 value.

238

Table 6.6.2-1: Summary of R2 for the Complete Data Set

SG (BULK) ACV (SSD) SG(SSD) SG(APP)
0.011

MSS

ACV

MSS LAA CFT ACV ACV (SSD) ABS SG (BULK) SG (SSD) SG (APP) PSF

0.536 0.116 0.320 0.220 0.196 0.401 0.172 0.114 0.012 0.037 0.179 0.389 0.124 0.080 0.536 0.311 0.247 0.079 0.013 0.041 0.650 0.487 0.075 0.171 0.180 0.168 0.000 0.012 0.005 0.147 0.031 0.020 0.009 0.004 0.836 0.141 0.167 0.161 0.122 0.033 0.134 0.172 0.169 0.133 0.028 0.647 0.492 0.140 0.004 0.975 0.744 0.009 0.854 0.011

According to statistical analysis, only R2 values above approximately 0.60 can indicate a notable correlation between two variables. Hence, values greater than 0.60 are

highlighted in Table 6.6.2-2. One of the graphs showing a correlation is aggregate crushing value test plotted against L.A. Abrasion test. The graph is shown in Figure 6.6.2-1.

239

ABS

LAA

CFT

MD

50 Aggregate Crushing Value, % Loss 45 40 35 30 25 20 15 10 5 0 10 20 30

y = 3.1264x 0.5835 R2 = 0.6495

40 LA Abrasion, % Loss

50

60

70

Figure 6.6.2-1:

L.A. Abrasion vs. Aggregate Crushing Value

As the graph demonstrates, a coefficient of determination of 0.65 was obtained using exponential regression analysis. A correlation was expected in this case since both tests measure aggregate strength to a certain extent aggregate crushing value test does this directly by subjecting the material to slow crushing pressure in a dry condition while the L.A. Abrasion test subjects the aggregate to repeated impact by large steel balls. However, statisticians consider a relationship strong when the R2 value is greater than approximately 0.75. Hence, the correlation is worth noting but is not very strong in this case. A strong linear correlation was observed when aggregate crushing value test was plotted against aggregate crushing value (saturated, surface dry). Figure 6.6.2-2

illustrates the strong relationship with the R2 value of 0.84. This behavior was expected since the two tests are very similar in their procedures with the only difference in the

240

60.00 Aggregate Crushing Value (SSD), % Loss 50.00 40.00 30.00 20.00 10.00 0.00 10

y = 0.9539x + 1.5506 R2 = 0.836

15

20

25

30

35

40

45

50

Aggregate Crushing Value, % Loss

Figure 6.6.2-2:

Aggregate Crushing Value (SSD) vs. Aggregate Crushing Value

fact that testing of the aggregate is done when the pores are saturated and the surface is dry in the latter case. A notable correlation was observed when absorption was plotted against bulk specific gravity as Figure 6.6.2-3 indicates. During the regression analysis, the highest possible value of R2 obtained was 0.65, which indicated that a relationship exists. However, statistically it is not a significant one since the value is less than 0.75, and hence it can be concluded that the correlation between the two tests is not strong.

241

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 1.0 2.0 3.0 Absorption 4.0 5.0 6.0

y = 2.7496e-0.0357x R2 = 0.6473

Figure 6.6.2-3:

Specific Gravity (Bulk) vs. Absorption

A strong correlation was expected when bulk specific gravity was plotted against saturated, surface-dry specific gravity. The plot is shown in Figure 6.6.2-4.
3.20 3.00 Specific Gravity (SSD) 2.80 2.60 2.40 2.20 2.00 2.20

y = 0.8356x + 0.4662 R2 = 0.9749

2.30

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (Bulk)

Figure 6.6.2-4:

Specific Gravity (SSD) vs. Specific Gravity (Bulk)

242

Regression analysis showed that the linear relationship produces the highest R2 value of 0.97. As the coefficient of determination indicates, the relationship is very strong, and hence, it can be recommended that the value of one test can be predicted using the equation of the line given in Figure 6.6.2-4. A similar behavior was observed when bulk specific gravity was plotted against apparent specific gravity. As Figure 6.6.2-5 indicates, the coefficient of determination of 0.74 is high and is achieved when quadratic curve is used. Thus, it can be concluded that
3.20 3.00 Specific Gravity (Apparent) 2.80 2.60 2.40 2.20 2.00 2.20

y = 0.4688x 2 - 1.8333x + 4.2933 R2 = 0.7437

2.30

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (Bulk)

Figure 6.6.2-5:

Specific Gravity (Apparent) vs. Specific Gravity (Bulk)

the relationship is strong, and the two tests can be correlated using the equation provided in Figure 6.6.2-5 with relatively high certainty. Another notable correlation was noted when apparent specific gravity was plotted against saturated, surface-dry gravity as shown in Figure 6.6.2-6. A coefficient of

determination of 0.85 was obtained when quadratic curve was used in the regression analysis. Using linear regression analysis produced similarly high value of the 243

coefficient of determination. The R2 value is close to 1.0 and hence the correlation is strong when the curve shown in Figure 6.6.2-6 is used to correlate the two tests.
3.20 3.00 Specific Gravity (Apparent) 2.80 2.60 2.40 2.20 2.00 2.20

y = 0.3415x 2 - 1.0461x + 3.0821 R2 = 0.8538

2.30

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

3.20

Specific Gravity (SSD)

Figure 6.6.2-6:

Specific Gravity (Apparent) vs. Specific Gravity (SSD)

Unfortunately, no other tests showed notably strong correlations. In particular, micro-Deval did not correlate well with any of the tests. Interestingly, an R2 value of only 0.12 was observed between L.A. Abrasion and micro-Deval tests indicating that no correlation exists between the two tests even though both tests subject the aggregates to impact by steel balls. A possible explanation for this behavior could lie in the fact that the aggregate pores are saturated with water in micro-Deval testing while the aggregate is completely dry when tested in the L.A. Abrasion machine. Furthermore, no correlation was observed between magnesium sulfate soundness and Canadian freeze-thaw tests despite the fact that both measure aggregate resistance to freeze-thaw cycles. One

possible explanation for such behavior lies in the extremely large test data variations observed during magnesium sulfate testing for each source. 244 Additionally, analysis

showed that the particle shape factor cannot be correlated to any of the tests carried out during this study as computed R2 values are extremely low.
6.6.3 Partial Data Set Regression Analysis I

During computations of R2 values, outliers carry an extremely negative impact. A presence of only a few outliers can reduce the value of the coefficient of determination significantly. Hence, using the assumption of normal population distribution, some

outliers can be eliminated by including points only within the range of 3 * , where is the data set mean and is the sample standard deviation. Using this procedure, 99.7% of the data points will still be included and some extreme outliers will be eliminating. Thus, a correlation will be more readily seen as the R2 values should be higher if a correlation exists but was previously disguised by the presence of outliers. Using Microsoft Excel 2003, the data for each test were analyzed and outliers eliminated as described above. A complete set of graphs was then produced and is included in Appendix R. Additionally, Appendix T includes Microsoft Excel tables showing which points were eliminated based on the prescribed criteria. Table 6.6.3-1 provides a summary of maximum R2 values obtained using linear, logarithmic, seconddegree polynomial, power, and exponential regression analysis techniques. The reader should refer to Appendix R for the equations of the best-fit curves that resulted in the highest possible R2 values.

245

Table 6.6.3-1: Summary of R2 for the Partial Data Set I

SG (BULK) ACV (SSD)

SG (SSD)

MSS LAA CFT ACV ACV (SSD) ABS SG (BULK) SG (SSD) SG (APP) PSF

0.565 0.122 0.223 0.171 0.154 0.330 0.211 0.190 0.132 0.029 0.129 0.393 0.074 0.032 0.528 0.279 0.218 0.088 0.001 0.045 0.704 0.571 0.063 0.151 0.163 0.135 0.006 0.011 0.015 0.120 0.031 0.019 0.009 0.007 0.748 0.011 0.068 0.095 0.091 0.014 0.036 0.148 0.116 0.120 0.021 0.623 0.503 0.147 0.064 0.978 0.722 0.041 0.831 0.040 0.043

As Table 6.6.3-1 demonstrates, no significant changes in R2 values were observed indicating that the very extreme outliers do no influence the data significantly. One R2 change worth noting did occur for the L.A. Abrasion test vs. aggregate crushing value test since the R2 value increased from 0.65 to 0.71 indicating that the relationship between the two tests is statistically important and should be noted.
6.6.4 Partial Data Set Regression Analysis II

Second data reduction was performed on including only the values that lie in the range of 2 * for each test, where once again is the sample mean and is the sample standard deviation. Using this method, 95% of the original data are included within calculations of R2 value. It should be noted, however, that upon performing data reduction, R2 values are not going to necessarily increase since it is possible that outliers positively contribute to the computed R2 value and removing them will reduce the 246

SG (APP)

MSS

ACV

ABS

LAA

CFT

MD

coefficient of determination.

The computations were carried out using linear,

logarithmic, second-degree polynomial, power, and exponential regression analysis techniques. The entire graph set is contained in Appendix S, and Microsoft Excel tables showing which points were eliminated to obtain the graphs are included in Appendix T. The results are summarized in Table 6.6.4-1.Avoid using

Table 6.6.4-1: Summary of R2 for the Partial Data Set II

SG (BULK) ACV (SSD)

SG (SSD)

MSS LAA CFT ACV ACV (SSD) ABS SG (BULK) SG (SSD) SG (APP) PSF

0.452 0.107 0.232 0.171 0.190 0.270 0.166 0.174 0.123 0.010 0.157 0.320 0.070 0.047 0.469 0.212 0.170 0.088 0.033 0.068 0.701 0.613 0.041 0.126 0.089 0.090 0.022 0.017 0.025 0.119 0.025 0.034 0.002 0.002 0.857 0.005 0.040 0.035 0.084 0.005 0.002 0.020 0.019 0.103 0.031 0.555 0.465 0.116 0.063 0.969 0.667 0.064 0.795 0.067 0.054

As the table indicates, the only notable improvement in the R2 value is evident for aggregate crushing value test (saturated, surface-dry) versus L.A. Abrasion test. However, the R2 value is only 0.61 indicating that the relationship is weak. Such behavior was expected since a significant relationship was observed between L.A. Abrasion and aggregate crushing value test, and aggregate crushing value test is closely related to the aggregate crushing value (SSD) test. 247

SG (APP)

MSS

ACV

ABS

LAA

CFT

MD

6.6.5

Results Summary

Statistical analyses were performed on test data to determine if correlations existed between any two tests performed in this research project regardless of performance. Consequently, two data reductions were performed in an attempt to find any correlations previously disguised by the presence of outliers and hence negatively impacting the computed R2 values. The following conclusions can be drawn: 1. Micro-Deval did not have statistically significant correlations to any other tests. 2. The following pairs of tests were found to have significant correlations: L.A. Abrasion and aggregate crushing value, L.A. Abrasion and aggregate crushing value (saturated, surface-dry), aggregate crushing value and aggregate crushing value (saturated, surface-dry), absorption and bulk specific gravity, bulk specific gravity and saturated surface-dry specific gravity, bulk specific gravity and apparent specific gravity, as well as saturated surface-dry specific gravity and apparent specific gravity. 3. Upon performing data reduction, the significance level of the following relationships increased: L.A. Abrasion and aggregate crushing value,

L.A. Abrasion and aggregate crushing value (SSD), as well as aggregate crushing value and aggregate crushing value (SSD). 4. Magnesium sulfate soundness test did not correlate well to Canadian freeze-thaw test since the maximum obtained R2 value was 0.39. 5. Particle shape factor test had extremely low R2 values and hence did not correlate at all to any of the tests carried out during this research.

248

6.7

Micro-Deval and AIMS Analysis

6.7.1 Background Information

As part of this research project, attempts were made to identify other methods that could improve the effectiveness of correctly classifying aggregates with respect to performance. Thus, a supplementary study was carried out on 20 selected aggregates with various performances using Aggregate IMaging System (AIMS). The system was primary developed at Texas A&M University by Dr. Eyad Masad, who was instrumental in carrying out testing and analysis of this phase of the research project. AIMS is a software-controlled test, which involves taking of black and white as well as gray scale images of aggregates as they pass under a camera. The camera can achieve various levels of magnification as well as move in x, y, and z directions. Specific procedures for carrying out the test can be found in Implementation Report 5-1707-01-1 published by Texas Transportation Institute. As the result, the system can capture all three dimensions of an aggregate and compute several crucial properties that characterize it form, angularity, and texture. Form characteristic is quantified through the use of sphericity index defined as Sphericity = 3
d s .d l d2 L

where ds is the shortest dimension, dl is the longest dimension, and di is the intermediate dimension. Angularity is characterized via the angularity index computed using the gradient method and is defined as Angularity Index =

N 1 i =1 3

N 3

i i +3

249

where N is the total number of aggregate boundary points and i denotes the ith point on the boundary. Finally, texture is characterized through the use of the texture index defined as
1 Texture Index = 3N

( D ( x, y ) )
i =1 j =1 i, j

where D is the wallet coefficient and i represents the three directions.

For more

information on the formulas and their derivations, the reader is encouraged to refer to Appendix U, which represents the Summary Report by Dr. Eyad Masad on the aggregates submitted to his research team for testing.
6.7.2 Testing Procedure

Twenty aggregates out of the suite of 117 sources were chosen by the research team to be submitted for AIMS analysis. They represented a wide range of performances 10 good performers, 6 fair performers, and 4 poor performers. To remove any bias on the part of the testing team, performance data was not supplied until after the entire testing process was completed. The following steps outline the testing procedure: 1. Aggregates were weighed as well as separated according to size. 2. Aggregates were submitted for AIMS analysis at Texas A&M University. 3. Aggregates were returned and tested in the micro-Deval machine at the University of Texas at Austin. 4. Aggregates were properly sieved out as per ASTM micro-Deval test specifications and submitted to Texas A&M University for additional post-testing AIMS analysis. 5. AIMS test data was received by this research team and performance details were shared with the AIMS research group, which performed further analysis utilizing the new performance information. 250

6.7.3

Results Discussion

For each aggregate tested, sphericity, texture index, and angularity index were computed both before and after micro-Deval testing in addition to percent changes in each. The reader should refer to Appendix U for the report summarizing the test data. Sphericity index is similar to the particle shape factor previously computed and discussed in this report. Upon complete data examination, it was determined that the particle shape factor was not a good performance predictor nor did it contribute any significant information to improve micro-Deval performance predictions. Similar results were

observed for the sphericity factor. Upon consultation with Dr. Masad, it was determined that such behavior is not unexpected as his research experience has shown that sphericity or particle shape factor is generally not a good indicator of performance. However, the

texture and angularity indices did prove to be good performance predictors as will be discussed in the following paragraphs. Particular attention was given to angularity index information. Figure 6.7.3-1 provides the plot of performance versus the average AIMS angularity index, where 1 represents good performance, 2 represents fair performance, and 3 represents poor performance. The figure also provides the distribution information via the bracketed short lines at the top of each column. The important aspect of the graph worth noticing is the fact that the average angularity index ranks the aggregates correctly the highest angularity average is observed for good performers, lower angularity average is seen for fair performers, and the lowest angularity average is observed for poor performers.

251

4000 3500 3000 AIMS Angularity Index 2500 2000 1500 1000 500 0 1 2 Performance Category 3 BMD AMD

Figure 6.7.3-1:

AIMS Angularity Index vs. Performance

AIMS texture index and performance relationship were examined next. Figure 6.7.3-2 provides the plot of AIMS texture index versus performance, where performance numbers have the same designations as before. The reader should note that the average texture index correctly ranks the performers only in the After Micro-Deval case. While the average texture index for fair performers is lower than that of good performers, the average texture index for poor performers is actually higher than that of fair performers though still lower than that of good performers. Upon further examination, it was determined that one of the aggregates in the poor category had exceptionally high texture and hence skewed the results. Figure 6.7.3-3 represents the same plot as before but with that aggregate removed. As the figure demonstrates, the aggregates are now correctly ranked in both cases before and after micro-Deval testing. 252

350

300

250 AIMS Texture Index

200 BMD AMD 150

100

50

0 1 2 Performance Category 3

Figure 6.7.3-2:
350

AIMS Texture Index vs. Performance for All Aggregates

Without Aggregate ON-10 300

250 AIMS Texture Index

200 BMD AMD 150

100

50

0 1 2 Performance Category 2

Figure 6.7.3-3:

AIMS Texture Index vs. Performance with One Aggregate Removed

253

Good ranking capabilities due to angularity and texture indices led to the assertion that the two combined could provide good performance prediction. Hence, Figure 6.7.34 demonstrates the results when AIMS angularity index is plotted against AIMS texture index before micro-Deval testing was carried out. It is believed that aggregates with high
4500

4000

3500

3000 AIMS Angylarity Index

2500

Good Fair Poor

2000

1500

1000

500

0 0 50 100 150 200 AIMS Texture Index 250 300 350 400

Figure 6.7.3-4:

AIMS Angularity Index vs. AIMS Texture Index Before Micro-Deval

texture and high angularity (upper right quadrant) should exhibit good performance while aggregates with low texture and low angularity (lower left quadrant) should exhibit poor performance in the field since better aggregate interlock is achieved for highly angular and textured aggregates. Aggregates falling into the other two quadrants should be given the benefit of the doubt as to how they will perform in the field. To test the validity of this assertion, the quadrant boundaries were varied in order to optimize the correct 254

aggregate qualification percentage. Through numerous trials, it was determined that angularity index of 2850 and texture index of 220 produced optimal results. Using this quadrant division, 75% of poor performers and 80% of good performers were identified correctly. Additionally, two of the six fair performers fell into the upper left quadrant. However, it should be noted that one fair performer fell into the poor quadrant while three fair and one poor aggregates fell into the good quadrant. Thus, overall the results are quite good Attempts were made to determine whether or not the assertion held true for AIMS analysis performed after micro-Deval testing. Figure 6.7.3-5 represents a plot similar to that discussed above but utilizing post-testing results. The results are similar to those
3000

2500

2000 AIMS Angylarity Index

1500

Good Fair Poor

1000

500

0 0 50 100 150 200 AIMS Texture Index 250 300 350 400

Figure 6.7.3-5:

AIMS Angularity Index vs. AIMS Texture Index Before Micro-Deval

255

previously observed. Once again, 75% of poor performers and 80% of good performers are classified correctly. Additionally, the angularity threshold decreased to 2000, and the texture threshold decreased to 140. As before, only two fair performers fell into one of the other two quadrants. Furthermore, only two fair performers fell into the good quadrant while one more (total of two) fair performers fell into the poor quadrant. Overall, however, the results are quite good and similar to those observed before microDeval testing.
6.7.4 Summary and Conclusions

AIMS analysis was performed on 20 aggregates used in this study. Analysis was carried out both before and after micro-Deval testing to observe if any additional information can be gained to correctly classify aggregates according to their field performance. It was concluded that AIMS was effective in correctly ranking the

aggregates using their average angularity index. Additionally, the average texture index correctly ranked the aggregates using post-micro-Deval test data but did not rank them correctly using pre-micro-Deval test data due to one aggregate with exceptionally high texture. Furthermore, AIMS was effective in correctly classifying good and poor

aggregates using angularity and texture indices on the same plot. However, the success rate of correctly classifying fair performers was low. Thus, it is the conclusion of this research team that results produced by AIMS analysis are quite promising for providing additional information to micro-Deval to correctly predict aggregate field performance. However, since only four poor aggregates were used and classification of the six fair aggregates was not effective, it is the recommendation of this research team that further studies need to be conducted with higher number of poor aggregates to definitively conclude just how effective micro-Deval and AIMS combination is in correctly classifying aggregate field performance. 256

Chapter 7: Summary and Conclusions

7.1 Project Summary

Due to excellent communication with numerous transportation agencies within the United States and Canada, an international study was possible and carried out to investigate the effectiveness of micro-Deval and other tests in successfully qualifying future field performance of an aggregate based on laboratory test data. During the material collection stage of the project, successful attempts were made to obtain aggregates representing diverse mineralogical backgrounds as well as various field performances. Upon reception of 117 sources, the following laboratory tests were carried out on each one: micro-Deval, magnesium sulfate soundness, L.A. abrasion,

Canadian freeze-thaw soundness, aggregate crushing value, aggregate crushing value (SSD), absorption, specific gravity (bulk, saturated surface dry, apparent), particle shape factor determination, and percent fractured test. Four major uses of aggregates were studied hot-mix asphalt, portland cement concrete, base course, and open-graded friction course. Within each use category,

prediction success rates were studied for micro-Deval alone as well as two-test combinations involving micro-Deval. Special attention was paid to aggregate mineralogy and climactic zones in which they were used. Furthermore, investigations into all other two-test combinations were carried out. Finally, a test correlation study was performed to determine if any correlations existed between the tests conducted during this research project in an attempt to determine if laboratory results of one test can be used to predict test results of another test.

257

7.2

Major Conclusions

Within Section 6, detailed conclusions were presented at the end of every major section. Taking that information into consideration, the following main conclusions can be drawn: 1. Micro-Deval proved to be a very consistent and reproducible test. 2. Laboratory comparison proved the wide variability of magnesium sulfate soundness test results. 3. Micro-Deval proved to be the best single-test indicator of performance for all four uses studied. 4. Notable classification improvements were obtained by combining microDeval test data with other test data. 5. In particular, Canadian freeze-thaw soundness, magnesium sulfate soundness, absorption, and specific gravity consistently yielded classification

improvements. 6. By separating aggregates into mineralogical groups, classification

improvements were obtained for siliceous gravel and granite subgroup but not for limestone and dolomite subgroup. 7. Inclusion of modified Thornthwaite index information yielded further improvements in correct aggregate classification for certain regions. 8. Addition of AIMS analysis information showed much promise for classification success rate improvement. 9. Regression analysis showed that micro-Deval did not correlate well to any other test carried out during this research project. 10. Statistically significant correlations were observed between the following pairs of tests: L.A. Abrasion and aggregate crushing value, L.A. Abrasion 258

and aggregate crushing value (saturated, surface-dry), aggregate crushing value and aggregate crushing value (saturated, surface-dry), absorption and bulk specific gravity, bulk specific gravity and saturated surface-dry specific gravity, bulk specific gravity and apparent specific gravity, as well as saturated surface-dry specific gravity and apparent specific gravity.
7.3 Need for Future Research

Two important aspects of this research project require further study that could result in gaining of additional useful information. The first aspect that requires further attention is the need to include more poor and fair sources within base course category. Only two fair and three poor sources were available in this study, and hence, prediction success rate comparison of micro-Deval to other tests alone was not possible. However, if more poor and fair sources can be obtained to achieve a balance between the numbers of fair, poor, and good aggregates, it will become possible to verify the patterns observed in other usage categories. The second aspect requiring further attention is the need to obtain greater field performance information from participating transportation agencies. For the purposes of this research study, field performance of aggregates was separated into three major categories of poor, fair, or good based on a formal survey created by ICAR 507 research team. While such category division was sufficient to successfully conduct the analysis, it is possible that better results or more useful information can be gained if the performance can be divided into more specific categories by dedicating a considerably greater amount of time investigating actual field performance based on a much more detailed criteria that would involve site visits and standardized pavement testing.

259

Appendix A:

Performance Questionnaire

261

ICAR 507: Aggregate Performance Questionaire

State: Supplier: Source: Mineralogy: USES Has this source been used in hot-mix asphalt (HMA)? How many years has the aggregate been in service? Average age before problems arose? What types of problems? Can these problems be traced directly to a failure of the aggregate? If Yes, what evidence suggests this? Was this an isolated event? What was/is the Average Annual Daily Traffic (AADT)? What was/is the environment? Exposure to FREEZETHAW? Has this source been used in Portland Cement Concrete (PCC)? How many years has the aggregate been in service? Average age before problems arose? What types of problems? Can these problems be traced directly to a failure of the aggregate? If Yes, what evidence suggests this? Was this an isolated event? What was/is the Average Annual Daily Traffic (AADT)? What was/is the environment?

ICAR 507 source ID

YES

NO

YES YES

NO NO

YES YES

NO NO

YES YES

NO NO

262

Exposure to FREEZETHAW? Has this source been used as a base course (BC)? Is it used as a bound and/or unbound base? How many years has the aggregate been in service? Average age before problems arose? What types of problems? Can these problems be traced directly to a failure of the aggregate? If Yes, what evidence suggests this? Was this an isolated event? What was/is the Average Annual Daily Traffic (AADT)? What was/is the environment? Exposure to FREEZETHAW? Has this source been used as an open-graded friction course (OGFC)? How many years has the aggregate been in service? Average age before problems arose? What types of problems? Can these problems be traced directly to a failure of the aggregate? If Yes, what evidence suggests this? Was this an isolated event? What was/is the Average Annual Daily Traffic (AADT)? What was/is the environment? Exposure to FREEZETHAW? Has this source been used for any other use? What? How many years has the aggregate been in service? Average age before problems arose?

YES YES

NO NO

YES YES

NO NO

YES YES

NO NO

YES YES

NO NO

YES YES

NO NO

263

What types of problems? Can these problems be traced directly to a failure of the aggregate? If Yes, what evidence suggests this? Was this an isolated event? What was/is the Average Annual Daily Traffic (AADT)? What was/is the environment? Exposure to FREEZETHAW? YES NO

YES

NO

YES

NO

Problem examples:

Pop outs Rutting Raveling Polishing

Fatigue cracking D-cracking Stripping

Degradation during mixing Degradation during construction Degradation under usage

The following rating system is to be used by ICAR 507 personnel, not by aggregate supplier, using information attained above. Please circle one. RATINGS Good Fair

Aggregate's life expectancy is 10 or more years with no reported non-chemical problems. Aggregate only experiences minor non-chemically related failures over its lifetime requiring minor repairs, but its average service life extends beyond 10 years Aggregate experiences severe degradation or failures within 2 years or service or during construction that severely inhibit and/or prevent the use of the application Good Good Good Good Good Fair Fair Fair Fair Fair Poor Poor Poor Poor Poor

Poor

HMA PCC BC OGFC Other:

264

Appendix B:

Micro-Deval Test Results

265

SAMPLE ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

GRADING

A A A A A A X X X A A A A A A A A A A A A A A A A A A B A B A A A A A A A A

1 17.2 18.3 29.7 22.3 9.7 10.8 X X X 13.4 28.0 4.7 7.4 9.1 12.4 9.0 8.5 9.0 17.2 27.2 7.4 13.8 8.9 9.7 6.4 7.6 48.6 19.1 14.7 3.0 21.4 21.7 7.8 12.1 11.4 16.1 19.3 22.2

TEST NUMBER 2 17.0 19.7 29.0 21.7 9.6 10.7 X X X 13.1 27.2 4.1 6.9 11.0 13.2 8.9 8.8 9.4 17.0 26.8 7.5 14.3 9.2 9.4 6.4 8.1 48.9 20.0 14.5 3.0 21.6 21.1 8.1 12.7 11.4 15.6 19.1 23.2

3 16.8 19.5 29.9 22.6 X 11.6 X X X 12.9 27.5 3.9 6.9 10.8 12.2 8.5 8.5 8.9 16.4 27.0 7.1 13.1 8.8 9.4 6.3 7.9 X X 15.2 3.3 21.4 21.5 7.5 12.8 12.2 15.1 19.4 22.0

AVG 17.0 19.2 29.5 22.2 9.7 11.0 X X X 13.1 27.6 4.2 7.1 10.3 12.6 8.8 8.6 9.1 16.9 27.0 7.3 13.7 9.0 9.5 6.4 7.9 48.8 19.6 14.8 3.1 21.5 21.4 7.8 12.5 11.7 15.6 19.3 22.5

VAR 0.04 0.57 0.22 0.21 0.01 0.24 X X X 0.06 0.16 0.17 0.08 1.09 0.28 0.07 0.03 0.07 0.17 0.04 0.04 0.36 0.04 0.03 0.00 0.06 0.05 0.40 0.13 0.03 0.01 0.09 0.09 0.14 0.21 0.25 0.02 0.41

266

39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81

A A A A A A A A A A A A A A A A A A A A A A A A X A A A A A A A A A A A A A A A A A A

10.3 23.2 19.3 19.0 4.6 14.0 21.5 10.5 13.7 14.5 17.7 21.7 1.7 1.4 17.3 7.0 17.4 21.7 34.8 3.9 23.7 3.3 8.2 27.0 X 8.2 20.7 23.4 10.0 5.4 9.9 13.7 12.2 13.8 7.2 9.5 22.0 31.0 11.8 17.5 15.1 25.6 18.2

10.8 22.5 20.2 19.2 4.5 13.6 21.4 9.0 13.1 14.9 16.4 20.6 1.6 1.7 16.3 7.3 17.9 20.8 34.3 4.5 23.6 4.1 9.5 26.7 X 8.0 21.6 23.2 9.8 5.1 10.3 13.6 12.5 14.4 7.3 9.5 23.1 31.3 12.0 18.1 13.8 27.8 18.7

10.4 23.2 X 18.9 4.2 14.1 22.1 9.3 13.1 14.9 16.7 22.4 1.4 1.1 15.6 7.5 16.5 19.3 33.7 3.5 25.3 X 7.3 27.0 X 7.8 20.7 22.8 9.9 6.0 10.6 13.6 12.3 13.6 7.7 9.3 20.4 30.3 12.5 19.0 14.3 28.9 16.3

10.5 23.0 19.8 19.0 4.4 13.9 21.7 9.6 13.3 14.8 16.9 21.6 1.6 1.4 16.4 7.3 17.3 20.6 34.3 4.0 24.2 3.7 8.3 26.9 X 8.0 21.0 23.1 9.9 5.5 10.3 13.6 12.3 13.9 7.4 9.4 21.8 30.9 12.1 18.2 14.4 27.4 17.7

0.07 0.16 0.40 0.02 0.04 0.07 0.14 0.63 0.12 0.05 0.46 0.82 0.02 0.09 0.73 0.06 0.50 1.47 0.30 0.25 0.91 0.32 1.22 0.03 X 0.04 0.27 0.09 0.01 0.21 0.12 0.00 0.02 0.17 0.07 0.01 1.84 0.26 0.13 0.57 0.43 2.82 1.60

267

82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117

A A A X A A A A C A A A A A A A A A A A A A A A A X B X A A A A A A A A

20.4 9.2 18.3 X 8.0 8.5 5.5 13.8 28.9 13.0 10.1 99.7 20.3 9.3 18.1 27.1 11.0 10.7 4.5 18.8 38.6 22.0 3.3 10.5 14.9 X 21.4 X 16.8 6.3 22.8 7.0 8.5 14.5 10.3 31.2

20.9 8.9 18.7 X 7.3 8.7 5.2 13.8 29.1 12.5 10.0 99.9 19.5 10.1 16.9 28.3 11.0 10.8 4.3 20.2 38.6 22.1 2.6 10.6 13.4 X 19.0 X 16.8 6.9 22.2 7.3 8.7 14.8 9.8 30.2

20.0 8.8 18.9 X 7.7 9.0 5.4 14.0 28.6 11.9 9.9 X 19.1 10.3 17.0 28.0 10.6 10.5 4.5 19.8 40.8 23.8 3.3 10.8 13.3 X 20.5 X 16.5 6.9 22.2 6.9 8.7 14.2 9.1 29.7

20.4 9.0 18.6 X 7.7 8.7 5.4 13.9 28.9 12.5 10.0 99.8 19.6 9.9 17.3 27.8 10.9 10.7 4.4 19.6 39.3 22.6 3.1 10.6 13.9 X 20.3 12.4 16.7 6.7 22.4 7.1 8.6 14.5 9.7 30.4

0.20 0.04 0.09 X 0.12 0.06 0.02 0.01 0.06 0.30 0.01 0.02 0.37 0.28 0.44 0.39 0.05 0.02 0.01 0.52 1.61 1.02 0.16 0.02 0.80 X 1.47 X 0.03 0.12 0.12 0.04 0.01 0.09 0.36 0.58

268

Appendix C:

Magnesium Sulfate Soundness Test Results

269

SAMPLE ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

TEST NUMBER 1 2 3 11.2 13.9 11.5 70.0 70.2 66.5 69.4 71.1 X 23.5 24.2 X 1.3 1.0 1.0 2.4 2.8 3.9 X X X X X X X X X 3.1 2.5 2.5 15.9 16.9 14.1 2.0 2.1 3.6 3.8 4.0 3.0 5.5 6.3 X 10.5 9.5 9.9 1.9 1.6 2.8 1.4 1.5 0.5 1.0 1.2 3.2 3.9 5.1 2.7 12.2 14.7 10.3 0.4 0.4 0.4 4.7 4.5 X 4.9 3.2 3.1 6.7 5.7 6.0 0.9 1.9 1.4 1.0 1.4 2.1 66.0 72.7 X X X X 11.5 14.1 13.1 0.6 0.6 1.0 5.6 7.1 5.2 13.2 8.2 10.0 7.9 10.8 8.7 4.9 5.8 3.6 2.0 1.4 2.4 2.1 1.8 1.9 5.8 5.8 3.7 6.6 7.9 9.7

Average 12.2 68.9 70.3 23.9 1.1 3.0 X X X 2.7 15.6 2.6 3.6 5.9 10.0 2.1 1.1 1.8 3.9 12.4 0.4 4.6 3.7 6.1 1.4 1.5 69.4 X 12.9 0.7 6.0 10.5 9.1 4.8 1.9 1.9 5.1 8.1

COV 2.2 4.3 1.4 0.2 0.0 0.6 X X X 0.1 2.0 0.8 0.3 0.3 0.3 0.4 0.3 1.5 1.4 4.9 0.0 0.0 1.0 0.3 0.3 0.3 22.4 X 1.7 0.1 1.0 6.4 2.2 1.2 0.3 0.0 1.5 2.4

270

39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81

3.9 0.4 9.4 22.0 0.5 6.5 4.6 6.7 6.5 7.3 9.8 22.5 7.4 2.7 3.9 1.8 2.4 1.0 21.2 2.6 39.7 2.0 9.6 26.9 X 7.6 15.1 8.7 4.7 0.8 3.0 4.0 1.7 11.5 1.5 1.2 6.2 43.3 5.9 17.3 5.9 18.6 22.1

2.8 0.5 8.6 26.1 0.3 6.4 4.4 6.6 10.1 12.7 7.7 24.8 6.8 3.9 5.3 1.8 2.5 1.8 18.8 2.6 36.8 2.7 9.7 29.2 X 7.1 14.5 6.8 4.1 0.6 1.7 4.7 1.0 12.6 0.9 1.2 5.7 35.5 5.9 16.5 5.6 16.7 19.0

3.6 0.9 X 17.9 0.2 6.2 2.5 7.3 8.3 9.9 8.5 24.0 5.7 4.0 5.6 0.9 3.1 1.1 X 4.3 33.7 1.6 X 28.7 X X X 6.6 3.7 0.7 5.5 4.8 0.8 X 0.6 0.9 X 38.3 X X 5.0 X 21.3

3.4 0.6 9.0 22.0 0.3 6.4 3.8 6.9 8.3 10.0 8.7 23.8 6.6 3.5 4.9 1.5 2.7 1.3 20.0 3.2 36.7 2.1 9.7 28.3 X 7.4 14.8 7.4 4.2 0.7 3.4 4.5 1.2 12.1 1.0 1.1 6.0 39.0 5.9 16.9 5.5 17.7 20.8

0.3 0.1 0.3 16.8 0.0 0.0 1.3 0.1 3.2 7.3 1.1 1.4 0.7 0.5 0.8 0.3 0.1 0.2 2.9 1.0 9.0 0.3 0.0 1.5 X 0.1 0.2 1.3 0.3 0.0 3.7 0.2 0.2 0.6 0.2 0.0 0.1 15.6 0.0 0.3 0.2 1.8 2.6

271

82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117

17.4 5.4 10.3 16.3 1.2 2.1 1.0 7.5 41.0 15.3 3.8 X 20.1 5.5 18.9 30.9 6.9 8.2 0.7 7.7 34.4 23.0 1.6 11.0 13.1 X 13.4 4.2 8.9 3.6 5.5 2.4 7.3 2.1 7.8 32.2

11.5 6.6 14.3 16.3 1.0 2.1 0.8 6.0 38.2 16.6 2.8 X 17.7 7.4 18.5 32.9 7.2 8.3 0.9 3.1 27.4 22.0 1.4 9.5 14.4 X 13.5 3.7 7.8 3.1 3.7 1.7 5.1 1.7 9.4 33.7

8.8 8.0 11.2 16.3 1.8 1.7 1.1 7.2 38.2 15.2 1.8 X 18.6 6.3 X 29.8 4.7 8.5 0.7 5.4 X 20.9 2.3 7.8 13.9 X X 2.7 8.2 3.8 5.9 1.7 5.2 X 10.5 X

12.6 6.7 11.9 16.3 1.3 2.0 1.0 6.9 39.1 15.7 2.8 X 18.8 6.4 18.7 31.2 6.3 8.3 0.8 5.4 30.9 22.0 1.8 9.4 13.8 X 13.5 3.5 8.3 3.5 5.0 1.9 5.9 1.9 9.2 33.0

19.3 1.7 4.4 0.0 0.2 0.1 0.0 0.6 2.6 0.6 1.0 X 1.5 0.9 0.1 2.5 1.9 0.0 0.0 5.3 24.5 1.1 0.2 2.6 0.4 X 0.0 0.6 0.3 0.1 1.4 0.2 1.5 0.1 1.8 1.1

272

Appendix D:

Los Angeles Abrasion Test Results

273

SAMPLE ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

GRADING

B B B B B B X X X B B B B B B B B B B B B B B B B B B B B B B B B B B B B B

1 30.9 65.7 57.0 30.1 21.0 26.3 X X X 13.6 12.9 20.2 14.2 14.5 31.5 18.0 17.5 26.7 13.7 32.1 15.2 58.0 61.8 57.8 36.9 51.7 33.9 24.4 34.8 14.5 27.0 31.0 19.1 23.9 21.2 13.2 41.6 34.3

TEST NUMBER 2 3 31.6 X 66.4 X 56.0 X 28.8 X 20.9 X 26.6 X X X X X X X 13.7 X 13.1 X 21.5 X 13.7 X 15.9 X 30.9 30.4 18.1 18.7 16.4 X 27.5 X 13.5 13.6 32.4 X 14.7 X 59.1 X 61.7 62.3 57.1 X 37.1 X 52.3 X 33.9 X 24.5 X 36.1 X 15.1 X 27.3 X 30.7 X 19.6 X 23.8 X 21.2 21.1 12.8 X 41.5 X 33.9 X

4 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

AVG 31 66 57 29 21 26 X X X 14 13 21 14 15 31 18 17 27 14 32 15 59 62 57 37 52 34 24 35 15 27 31 19 24 21 13 42 34

VAR 0.24 0.24 0.50 0.85 0.01 0.04 X X X 0.00 0.02 0.84 0.12 0.98 0.30 0.14 0.61 0.32 0.01 0.05 0.13 0.61 0.10 0.24 0.02 0.18 0.00 0.01 0.84 0.18 0.05 0.04 0.13 0.00 0.00 0.08 0.01 0.08

274

39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81

B B B B B B B B B B B B B B B B B B B B B B B B X B B B B B B B B B B B B B B B B B B

45.9 16.2 23.8 34.4 17.4 25.0 29.3 19.7 24.4 24.9 24.4 28.7 16.1 16.5 16.2 35.9 17.2 20.5 45.5 17.5 27.3 14.9 21.7 31.8 X 44.2 22.5 31.5 22.2 24.1 12.2 10.5 23.2 16.2 13.8 19.1 17.4 25.5 18.2 24.3 22.4 27.2 26.9

43.8 16.0 23.1 34.8 17.8 24.5 29.2 18.8 25.0 24.0 24.6 29.6 16.2 16.3 16.3 35.6 17.3 21.0 44.4 17.9 27.3 14.5 22.5 32.8 X 40.9 22.0 30.9 20.9 24.0 12.3 10.6 23.6 16.2 14.5 18.7 17.5 25.9 18.1 25.0 22.5 26.7 27.6

X X X X X 24.4 X 20.3 24.6 X X 28.6 X X X X X X X X X X X X X X X X X X X 10.9 X X 14.0 X X X X X X X X

X X X X X 24.6 X X 24.7 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

45 16 23 35 18 25 29 20 25 24 25 29 16 16 16 36 17 21 45 18 27 15 22 32 X 43 22 31 22 24 12 11 23 16 14 19 17 26 18 25 22 27 27

2.21 0.02 0.24 0.08 0.08 0.07 0.01 0.57 0.06 0.40 0.02 0.30 0.00 0.02 0.01 0.05 0.01 0.13 0.60 0.08 0.00 0.08 0.32 0.50 X 5.45 0.13 0.18 0.85 0.01 0.01 0.04 0.08 0.00 0.13 0.08 0.01 0.08 0.00 0.25 0.01 0.13 0.24

275

82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117

B B B B B B B B B B B B B B B B B B B B B B B B B X B B B B B B B B B B

31.6 21.5 26.6 24.1 33.4 37.9 13.3 23.8 36.1 36.0 26.3 59.3 31.7 22.1 26.4 18.0 23.2 31.0 22.3 51.4 55.8 29.3 18.1 33.8 26.4 X 34.0 23.3 30.1 19.3 27.9 33.7 27.2 22.6 24.4 34.7

31.9 22.3 25.6 23.1 33.2 37.1 13.7 22.0 35.5 37.1 25.7 61.2 31.9 22.7 25.6 18.3 23.9 31.2 23.3 52.5 56.4 28.7 18.8 34.2 26.6 X 34.0 23.5 30.4 19.8 27.5 34.1 28.1 21.7 25.8 34.9

X X X X X X X X X X X X X X X X X X X X X 29.1 18.5 34.0 X X X X X X X X X 21.8 X X

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

32 22 26 24 33 38 14 23 36 37 26 60 32 22 26 18 24 31 23 52 56 29 18 34 27 X 34 23 30 20 28 34 28 22 25 35

0.05 0.32 0.50 0.50 0.02 0.32 0.08 1.62 0.18 0.61 0.18 1.81 0.02 0.18 0.32 0.05 0.25 0.02 0.50 0.60 0.18 0.09 0.12 0.04 0.02 X 0.00 0.02 0.05 0.12 0.08 0.08 0.41 0.24 0.98 0.02

276

Appendix E:

Canadian Freeze-Thaw Test Results

277

TEST NUMBER SAMPLE ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 1 6.9 2.9 4.8 5.7 2.2 2.5 X X X 2.2 8.0 2.4 1.8 5.9 3.2 2.5 1.8 1.7 6.7 4.6 2.0 2.1 1.9 1.9 0.8 1.2 20.6 4.4 2.6 2.2 2.3 1.3 5.0 1.9 1.5 5.5 3.3 2.0 2 7.1 3.3 4.9 5.8 2.3 3.1 X X X 2.6 10.5 2.5 1.9 6.1 3.7 2.6 2.1 2.3 7.5 6.6 2.1 2.7 2.6 2.1 1.4 1.9 22.7 5.5 3.4 2.4 2.3 1.6 6.3 2.9 1.7 6.1 3.5 3.1 3 X X X X 2.3 X X X X X X X X X X X X X X X X X X X X X 22.9 X X X X X X X X X X X 4 X X X X 2.7 X X X X X X X X X X X X X X X X X X X X X 23.3 X X X X X X X X X X X Average 7.0 3.1 4.9 5.8 2.4 2.8 X X X 2.4 9.3 2.5 1.9 6.0 3.5 2.6 2.0 2.0 7.1 5.6 2.1 2.4 2.3 2.0 1.1 1.6 22.4 5.0 3.0 2.3 2.3 1.5 5.7 2.4 1.6 5.8 3.4 2.6 COV 0.0 0.1 0.0 0.0 0.0 0.2 X X X 0.1 3.1 0.0 0.0 0.0 0.1 0.0 0.0 0.2 0.3 2.0 0.0 0.2 0.2 0.0 0.2 0.2 1.5 0.6 0.3 0.0 0.0 0.0 0.8 0.5 0.0 0.2 0.0 0.6

278

39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81

1.4 1.1 4.5 2.0 1.4 8.5 2.0 10.1 6.0 7.0 9.0 7.3 3.6 3.4 5.1 1.0 1.1 1.8 1.5 3.2 13.9 2.5 3.9 9.7 X 1.5 5.9 6.1 2.1 5.7 2.9 3.1 2.9 3.3 1.2 3.1 3.2 6.0 7.2 9.8 9.8 7.4 3.2

1.9 1.4 4.7 2.8 1.7 9.4 3.0 10.7 7.6 8.0 9.1 10.3 4.1 3.8 6.9 1.0 1.5 2.9 4.6 3.8 15.6 3.0 4.0 10.7 X 4.2 6.6 6.6 2.6 5.8 3.3 3.4 4.6 3.7 1.7 4.3 3.8 7.2 9.5 14.4 10.7 9.0 4.0

X X X X X X X X X X X X X X X X X X X X X 3.3 X X X X X X X X X X X X X X X X X X X X X

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

1.7 1.3 4.6 2.4 1.6 9.0 2.5 10.4 6.8 7.5 9.1 8.8 3.9 3.6 6.0 1.0 1.3 2.4 3.1 3.5 14.8 2.9 4.0 10.2 X 2.9 6.3 6.4 2.4 5.8 3.1 3.3 3.8 3.5 1.5 3.7 3.5 6.6 8.4 12.1 10.3 8.2 3.6

0.1 0.0 0.0 0.3 0.0 0.4 0.5 0.2 1.3 0.5 0.0 4.5 0.1 0.1 1.6 0.0 0.1 0.6 4.8 0.2 1.4 0.2 0.0 0.5 X 3.6 0.2 0.1 0.1 0.0 0.1 0.0 1.4 0.1 0.1 0.7 0.2 0.7 2.6 10.6 0.4 1.3 0.3

279

82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117

4.1 2.4 4.4 10.3 1.6 0.9 1.0 4.3 9.0 4.1 2.2 99.0 12.5 2.3 8.2 13.4 2.9 4.6 0.9 1.7 3.0 10.3 0.5 0.9 3.1 X 1.6 3.8 3.7 0.8 2.7 0.7 2.4 4.5 2.8 4.8

4.4 2.6 X 10.8 1.8 1.7 1.8 4.7 11.2 4.3 4.5 99.5 13.3 3.1 8.6 14.0 2.9 5.0 1.3 2.1 4.2 11.3 0.6 1.3 3.4 X 2.3 4.0 4.2 0.9 3.0 1.8 2.6 4.9 4.3 6.4

X X X X X X X X X X X X X X X X X X X X X X X X X X X 4.5 X X X X X X X X

X X X X X X X X X X X X X X X X X X X X X X X X X X X 4.8 X X X X X X X X

4.3 2.5 4.4 10.6 1.7 1.3 1.4 4.5 10.1 4.2 3.4 99.3 12.9 2.7 8.4 13.7 2.9 4.8 1.1 1.9 3.6 10.8 0.6 1.1 3.3 X 2.0 4.3 4.0 0.9 2.9 1.3 2.5 4.7 3.6 5.6

0.0 0.0 #DIV/0! 0.1 0.0 0.3 0.3 0.1 2.4 0.0 2.6 0.1 0.3 0.3 0.1 0.2 0.0 0.1 0.1 0.1 0.7 0.5 0.0 0.1 0.0 X 0.2 0.2 0.1 0.0 0.0 0.6 0.0 0.1 1.1 1.3

280

Appendix F:

Aggregate Crushing Value Test Results

281

TEST NUMBER SAMPLE ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37

GRADING

3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8

1 20.5 27.1 25.0 22.1 20.4 20.5 X X X 14.1 16.1 16.0 12.3 13.9 24.6 18.0 15.5 22.7 15.7 28.7 12.4 29.2 31.0 32.2 23.9 26.3 25.7 45.4 25.6 14.9 30.0 25.0 16.4 19.0 18.4 13.5 26.1

2 19.4 27.4 25.0 21.6 19.5 20.6 X X X 14.7 15.1 14.8 11.4 13.9 23.9 17.8 15.3 22.5 15.7 28.7 13.0 27.6 32.0 31.6 23.7 26.6 25.4 45.7 26.9 14.3 30.4 26.5 16.1 19.1 18.5 13.0 26.3

3 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

4 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

ACV 20 27 25 22 20 21 X X X 14 16 15 12 14 24 18 15 23 16 29 13 28 32 32 24 26 26 46 26 15 30 26 16 19 18 13 26

282

38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80

3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8

23.1 29.4 15.9 19.2 25.7 17.0 23.5 25.3 20.5 23.5 23.0 22.5 22.1 10.3 12.8 16.8 26.8 16.1 19.4 40.4 12.9 20.2 15.1 15.4 24.7 30.7 21.4 29.8 21.7 X 11.8 14.8 20.0 12.8 15.2 18.5 18.2 21.3 19.6 24.7 22.0 18.8

22.3 30.1 15.5 20.1 24.9 16.5 23.1 25.0 20.3 23.3 23.8 22.2 21.7 10.7 12.7 16.1 27.2 16.4 19.3 40.8 12.2 21.4 14.0 15.9 24.0 31.5 21.2 30.2 21.6 X 13.6 14.7 21.4 12.2 15.7 20.5 16.4 20.3 20.0 24.7 22.6 19.0

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

23 30 16 20 25 17 23 25 20 23 23 22 22 11 13 16 27 16 19 41 13 21 15 16 24 31 21 30 22 X 13 15 21 13 15 20 17 21 20 25 22 19

283

81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117

3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8

21.6 21.9 12.2 17.5 22.6 27.6 28.2 13.2 20.3 23.7 24.6 26.2 32.9 26.5 19.7 22.0 16.5 23.1 23.5 17.3 26.7 48.1 23.8 16.5 21.2 17.0 X 18.3 22.1 21.8 13.9 20.0 22.7 23.0 19.4 16.9 23.4

21.1 22.5 12.3 20.2 X 28.0 27.7 12.8 20.4 23.3 25.0 25.3 33.6 26.6 20.5 21.7 15.3 23.6 23.3 17.8 26.0 48.7 23.7 17.9 21.4 17.3 X 17.7 22.5 22.5 12.5 19.8 21.9 22.1 19.7 16.8 22.8

X X X X X X X X X X X X X X 18.7 23.0 X X X X X X X X X X X X X X X X X X X X X

X X X X X X X X X X X X X X 18.3 21.9 X X X X X X X X X X X X X X X X X X X X X

21 22 12 19 23 28 28 13 20 24 25 26 33 27 19 22 16 23 23 18 26 48 24 17 21 17 X 18 22 22 13 20 22 23 20 17 23

284

Appendix G:

Aggregate Crushing Value (SSD) Test Results

285

TEST NUMBER SAMPLE ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

GRADING

3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8

1 19.4 30.5 26.7 24.8 19.5 22.8 X X X 16.2 19.3 16.7 14.3 15.5 24.3 19.5 17.0 23.0 15.8 28.5 14.6 31.0 31.7 32.8 25.0 26.9 24.3 49.9 7.2 15.0 29.1 27.1 18.1 20.0 20.0 14.9 25.4 23.2 30.8 17.6 18.5 23.3

2 19.4 29.5 26.1 24.2 19.8 25.0 X X X 16.5 18.1 16.5 13.6 16.1 24.1 20.1 17.6 22.1 17.5 27.9 14.9 31.1 30.8 33.1 26.2 29.7 24.2 50.4 8.0 16.1 25.4 26.2 18.9 20.7 20.5 16.8 26.8 22.9 31.5 17.0 17.1 22.3

3 X X X X X X X X X X X 15.1 13.6 X X X 15.9 22.6 16.8 X X X X X X X X X X X X X X X X 15.9 25.6 X X X X X

4 X X X X X X X X X X X 16.0 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

ACV 19 30 26 25 20 24 X X X 16 19 16 14 16 24 20 17 23 17 28 15 31 31 33 26 28 24 50 8 16 27 27 19 20 20 16 26 23 31 17 18 23

286

43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90

3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 1/2" 3/8+1/2" 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 1/2" 3/8 3/8 1/2" 3/8 3/8 3/8+1/2" 3/8 3/8 3/8 3/8 3/8 3/8

17.3 22.4 24.3 23.1 21.9 25.1 20.1 20.2 10.5 10.8 20.0 27.6 18.5 23.4 44.4 12.1 19.7 13.2 17.5 24.5 X 32.1 19.0 31.9 22.2 15.8 14.0 13.9 20.2 14.7 16.7 20.9 15.5 22.4 19.2 22.4 24.5 16.3 24.6 23.9 14.4 23.8 21.6 27.1 29.0 15.5 21.5 24.2

16.9 21.2 23.8 22.5 20.7 24.5 21.2 19.3 10.0 12.3 19.6 28.0 18.8 23.3 44.4 14.0 19.4 12.3 17.8 21.7 X 32.5 19.4 33.1 22.8 18.0 16.1 13.2 19.9 15.1 17.0 18.7 14.9 23.2 20.2 22.0 25.2 18.7 24.6 21.0 15.5 23.5 23.2 27.7 28.9 14.6 20.4 24.1

X 22.2 25.1 X X X X X X X X X X X X X X X X X X X X X X X X X X X 18.6 19.5 14.7 X 20.9 X X X X X X X X X X X X X

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 18.3 20.1 X X 20.2 X X X X X X X X X X X X X

17 22 24 23 21 25 21 20 10 12 20 28 19 23 44 13 20 13 18 23 X 32 19 33 23 17 15 14 20 15 18 20 15 23 20 22 25 18 25 22 15 24 22 27 29 15 21 24

287

91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117

3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8

25.2 25.6 X 29.9 19.7 22.8 19.9 26.7 26.0 22.0 33.2 47.2 23.6 15.8 20.2 17.6 X 13.7 21.4 21.2 12.9 21.4 26.6 24.5 17.5 15.7 22.2

24.2 26.0 X 30.7 19.5 22.1 20.8 26.2 27.3 21.5 31.1 42.9 23.2 14.8 22.6 17.0 X 9.0 20.4 19.0 14.9 21.8 23.2 24.5 20.4 15.7 21.4

X X X X X X X X X X X 48.5 X X X X X 20.1 X X X X X X X X X

X X X X X X X X X X X 54.2 X X X X X X X X X X X X X X X

25 26 X 30 20 22 20 26 27 22 32 48 23 15 21 17 X X 21 20 14 22 25 25 19 16 22

288

Appendix H:

Absorption and Specific Gravity Test Results

289

TEST RESULT SAMPLE ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

GRADING

SGBULK 2.58 2.44 2.43 2.51 2.75 2.67 X X X 2.76 2.83 2.57 2.70 2.72 2.59 2.76 2.76 2.63 2.59 2.36 2.82 2.63 2.61 2.65 2.61 2.63 2.45 X 2.60 2.62 2.32

SGSSD 2.63 2.49 2.50 2.57 2.76 2.69 X X X 2.79 2.89 2.60 2.73 2.74 2.62 2.77 2.77 2.65 2.61 2.44 2.83 2.65 2.63 2.66 2.63 2.64 2.57 X 2.62 2.63 2.40

SGAPP 2.70 2.58 2.62 2.66 2.79 2.72 X X X 2.86 3.01 2.63 2.78 2.77 2.65 2.80 2.79 2.67 2.65 2.56 2.84 2.69 2.66 2.68 2.65 2.66 2.79 X 2.67 2.66 2.52

ABS 1.6 2.4 2.9 2.2 0.6 0.7 X X X 1.3 2.2 0.8 1.1 0.7 0.8 0.5 0.3 0.5 0.9 3.4 0.3 0.8 0.7 0.5 0.5 0.5 5.0 X 1.0 0.6 3.3

3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8

290

32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67

3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8

2.42 2.71 2.59 3.01 2.66 2.88 2.67 2.61 2.90 2.66 2.53 2.73 2.65 2.59 2.69 2.67 2.64 2.64 2.59 2.40 2.51 2.79 2.62 2.88 2.75 2.27 2.65 2.41 2.60 2.65 2.49 X 2.63 2.65 2.26 2.63

2.52 2.75 2.64 3.02 2.67 2.88 2.69 2.62 2.91 2.69 2.60 2.74 2.67 2.61 2.73 2.69 2.67 2.66 2.65 2.48 2.55 2.80 2.63 2.87 2.77 2.37 2.67 2.51 2.62 2.69 2.56 X 2.69 2.67 2.39 2.65

2.67 2.82 2.72 3.05 2.69 2.90 2.72 2.65 2.94 2.74 2.72 2.75 2.71 2.66 2.79 2.72 2.71 2.74 2.75 2.60 2.62 2.84 2.65 2.86 2.79 2.53 2.72 2.67 2.66 2.75 2.68 X 2.79 2.71 2.60 2.68

3.8 1.5 1.8 0.3 0.4 0.3 0.7 0.7 0.5 1.1 2.7 0.3 0.8 0.9 1.3 0.8 1.0 1.0 2.2 3.1 1.8 0.7 0.4 0.2 0.4 4.5 1.0 4.0 0.8 1.3 2.9 X 2.2 0.8 5.7 0.7

291

68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103

3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8

2.67 2.55 2.66 2.75 2.58 2.88 2.72 2.62 2.49 2.70 2.69 2.68 2.49 2.47 2.47 2.63 2.61 2.67 2.66 2.66 2.73 2.60 2.39 2.63 2.82 X 2.67 2.64 2.32 2.68 2.52 2.64 2.61 2.74 2.29 2.54

2.68 2.60 2.71 2.77 2.62 2.90 2.72 2.66 2.57 2.71 2.70 2.70 2.58 2.57 2.56 2.68 2.65 2.68 2.67 2.67 2.74 2.64 2.45 2.66 2.83 X 2.69 2.67 2.43 2.73 2.58 2.66 2.63 2.76 2.40 2.59

2.69 2.67 2.81 2.79 2.70 2.96 2.74 2.73 2.69 2.73 2.73 2.73 2.75 2.73 2.72 2.76 2.73 2.71 2.69 2.69 2.74 2.70 2.54 2.70 2.84 X 2.72 2.74 2.61 2.81 2.68 2.69 2.66 2.79 2.56 2.69

0.2 1.8 2.1 0.5 1.7 0.9 0.3 1.5 2.9 0.4 0.6 0.6 3.7 3.8 3.7 1.7 1.8 0.6 0.5 0.5 0.1 1.4 2.5 0.9 0.3 X 0.8 1.3 4.7 1.7 2.3 0.6 0.7 0.7 4.7 2.2

292

104 105 106 107 108 109 110 111 112 113 114 115 116 117

3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8 3/8

2.63 2.57 2.51 X 2.21 2.67 2.64 2.51 2.72 2.62 2.60 2.62 2.53 2.56

2.64 2.60 2.56 X 2.26 2.69 2.69 2.55 2.74 2.64 2.62 2.65 2.57 2.60

2.65 2.65 2.65 X 2.33 2.72 2.76 2.62 2.77 2.67 2.67 2.71 2.64 2.67

0.3 2.0 2.0 X 2.4 0.6 1.6 1.6 0.7 0.7 1.0 1.3 1.7 1.5

293

Appendix I:Particle Shape Factor Test Results

295

SAMPLE ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

GRADING

TEST RESULTS L/W W/T L/T 1.47 1.59 1.59 1.72 1.62 1.51 X X X 1.56 1.50 1.50 1.45 1.50 1.57 1.63 1.55 1.43 1.74 1.45 1.42 1.23 1.37 1.51 1.39 1.42 1.68 1.45 1.54 1.28 1.79 1.58 1.83 1.55 2.14 2.10 2.24 X X X 1.86 1.66 1.57 1.66 2.11 2.03 2.51 2.18 1.52 2.52 1.69 1.88 1.91 2.21 1.82 1.78 1.46 2.48 2.34 2.22 1.54 2.21 2.27 2.46 2.37 3.00 2.86 3.11 X X X 2.73 2.38 2.19 2.35 2.72 2.69 3.25 2.79 2.06 3.43 2.25 2.50 2.29 2.63 2.38 2.35 2.12 3.31 2.87 2.89 1.91 3.04

PSF Factor 1.66 1.78 1.57 2.17 2.29 3.06 X X X 2.09 1.76 1.53 1.86 2.55 2.22 3.07 2.53 1.53 2.86 1.81 2.33 2.91 3.10 1.90 2.17 1.54 2.91 3.19 2.71 1.80 2.10

1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2

296

32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67

1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 3/8 1/2 1/2 1/2

1.72 1.65 1.40 1.47 1.80 1.67 1.47 1.53 1.59 1.53 1.54 1.67 1.66 1.62 1.75 1.51 1.68 1.61 1.52 1.78 1.54 1.55 1.67 2.08 2.03 1.57 1.49 1.40 1.59 1.37 1.45 X 1.71 1.52 1.55 1.74

2.59 2.18 1.81 1.89 2.57 2.35 1.60 2.09 2.54 2.42 1.56 2.23 2.29 2.29 2.36 1.59 2.10 2.07 1.88 2.20 1.86 2.44 2.05 2.81 2.85 2.10 1.79 1.67 2.09 1.66 1.88 X 2.11 1.82 1.79 2.22

3.46 2.95 2.22 2.70 3.65 3.26 2.20 2.75 3.18 3.59 2.22 2.98 3.05 3.05 3.34 2.20 2.91 2.78 2.46 3.50 2.70 3.17 2.73 4.23 4.18 2.72 2.45 2.12 2.74 2.17 2.68 X 2.99 2.57 2.37 3.05

3.03 2.36 2.05 2.36 2.90 2.75 1.63 2.46 3.19 3.71 1.46 2.38 2.53 2.66 2.57 1.53 2.17 2.22 2.00 2.43 2.12 3.22 2.01 2.75 2.89 2.32 1.98 1.81 2.27 1.92 2.40 X 2.16 2.02 1.77 2.24

297

68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103

1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2

1.89 1.48 1.35 1.75 1.47 1.51 1.44 1.36 2.23 1.44 1.64 1.68 1.37 1.43 1.43 1.43 1.50 1.73 1.73 1.68 1.97 1.73 1.70 1.58 1.80 X 1.61 1.67 1.52 1.61 1.72 1.63 1.48 1.35 1.49 1.60

2.25 1.86 1.74 2.10 1.72 1.92 1.71 1.76 2.77 1.51 1.82 2.05 1.66 1.89 1.72 1.56 1.82 2.35 2.18 2.03 2.64 2.26 1.96 2.75 2.09 X 2.08 1.86 2.32 2.32 2.31 1.90 1.79 1.81 2.29 2.94

3.37 2.67 2.23 3.13 2.37 2.82 2.37 2.28 4.13 2.09 2.77 2.84 2.12 2.66 2.38 2.18 2.56 3.18 3.03 2.89 3.72 3.12 2.72 3.47 3.01 X 2.75 2.64 2.99 3.06 3.20 2.60 2.48 2.51 3.31 3.72

2.12 2.27 2.13 2.15 1.89 2.37 1.95 2.17 2.30 1.52 1.87 2.06 1.88 2.46 2.00 1.66 2.07 2.50 2.21 2.08 2.53 2.36 1.84 3.82 1.94 X 2.21 1.76 3.00 2.74 2.50 1.86 2.03 2.49 3.41 4.27

298

104 105 106 107 108 109 110 111 112 113 114 115 116 117

1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2

1.61 1.23 1.50 X 1.57 1.48 1.48 1.44 1.57 1.33 1.72 1.50 1.53 1.70

1.75 1.55 3.12 X 2.03 1.93 1.66 1.89 2.55 1.69 2.12 1.75 2.42 2.52

2.72 1.90 3.74 X 2.49 2.43 2.29 2.55 3.78 2.22 2.88 2.44 3.14 3.47

1.84 1.95 5.19 X 2.05 2.14 1.74 2.32 3.91 2.12 2.06 1.90 3.25 3.03

299

Appendix J:

Percent Fractured Particles Test Results

301

Source 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

size 3/8" 1/2" 1/2" 1/2" 1/2" 1/2" X X X 1/2" 1/2" 3/8" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2"

# of Particles 2 or more 1 88 3 X X X X X X X X X X X X X X X X X X X X 95 4 85 8 61 19 64 0 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 73 11 3 7 X X 107 0 91 0 X X

none 6 X X X X X X X X X X 6 20 36 17 X X X X X X X X X X X X X X X X X 31 125 X 7 3 X

% Fractured 1+ 2+ 94 91 100 100 100 100 100 100 100 100 100 100 X X X X X X 100 100 100 100 94 90 82 75 69 53 79 79 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 73 63 7 2 100 100 94 94 97 97 100 100

302

39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81

1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" X 1/2" 1/2" 1/2" 1/2" X 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2"

X X 75 X X X X X X X 114 X 2 3 X X X X X 31 95 97 32 X X X X X 131 X X X X X X X 102 X X X X 83 X

X X 2 X X X X X X X 0 X 7 4 X X X X X 4 3 12 6 X X X X X 4 X X X X X X X 9 X X X X 6 X

X X 22 X X X X X X X 5 X 84 111 X X X X X 53 7 8 90 X X X X X 3 X X X X X X X 12 X X X X 36 X

100 100 78 100 100 100 100 100 100 100 96 100 10 6 100 100 100 100 100 40 93 93 30 100 X 100 100 100 98 X 100 100 100 100 100 100 90 100 100 100 100 71 100

100 100 76 100 100 100 100 100 100 100 96 100 2 3 100 100 100 100 100 35 90 83 25 100 X 100 100 100 95 X 100 100 100 100 100 100 83 100 100 100 100 66 100

303

82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117

1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 3/8" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" X 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2" 1/2"

X 8 X X X X 55 81 X X X X X 64 94 X 121 X X X X 19 X X X X X X X 114 X X 114 X 7 X

X 0 X X X X 2 3 X X X X X 6 4 X 3 X X X X 16 X X X X X X X 2 X X 2 X 4 X

X 99 X X X X 32 23 X X X X X 32 28 X 2 X X X X 62 X X X X X X X 6 X X 0 X 99 X

100 7 100 100 100 100 64 79 100 100 100 100 100 69 78 100 98 100 100 100 100 36 100 100 100 X 100 100 100 95 100 100 100 100 10 100

100 7 100 100 100 100 62 76 100 100 100 100 100 63 75 100 96 100 100 100 100 20 100 100 100 X 100 100 100 93 100 100 98 100 6 100

304

Appendix K:

Petrographic Analysis

305

ID # 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

Petrographic Analysis Crushed Siliceous & Limestone Gravel (Granite, Sandstone, Limestone) Quartz Sandstone Crushed Quartz Sandstone Quartzite Limestone Crushed Granite Quartzite Siliceuous (Chert) Gravel Granodiorite/ Quartzite Crushed Basalt Crushed Basalt Crushed Siliceous Gravel (chert, quartzite) Crushed Siliceous Gravel (basalt, rhyolite, granite) Siliceous (Igneous) Gravel Granite Crushed Biodite Granite Gneiss Crushed Biodite Granite Gneiss Crushed Granite Crushed Serpentinite Crushed Fossil Limestone Crushed Fine Grained Dolomite Crushed Biotite Granite Gneiss Granite Granite Crushed Biotite Granite Crushed Fine Grained Granite Dolomite/ Limestone Kiowa expanded Shale (light weight) Calcareous(?) Sandstone Crushed Granite Crushed Limestone Crushed Dolomite & Limestone Crushed & Uncrushed Siliceous, Limestone, & Dolomite Gravel Crushed & Partly Crushed Dolomite Gravel with some Crushed Granite Amphibolite/ Gneiss Complex Serpentinite Dolomite Crushed Marble

306

39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80

Granite Schist Partly Crushed Siliceous and Limestone Gravel Crushed Dolomite with Quartz Crushed Granite & Basalt Crushed Limestone Crushed Limestone Crushed Dolomite Crushed Limestone Crushed Limestone Crushed Limestone Crushed Limestone Siliceous Gravel (chert, quartzite) Siliceous Gravel (chert) Shale/ Slate Crushed Granite Crushed Phyllite Crushed Phyllite Crushed Fossiliferous Limestone with embedded Silica Sandy Limestone Siliceous and Limestone Gravel Crushed Sandy Limestone and Dolomite Quartzite Gravel Quartzite and Granite Gravel Crushed Limestone Gravel Basalt Limestone Caliche Limestone Gravel Quartzite Crushed Rhyolite (Granite) Crushed Basalt (Granite) Crushed Dolomite Crushed Dolomite and Dolomitic Limestone Gravel Crushed Diorite (or Andesite) Crushed Limestone Crushed Siliceous and Dolomite Gravel Siltstone/ Sandstone Crushed Limestone Crushed Limestone Crushed Limestone Siliceous Sandstone & Siliceous Limestone Gravel

307

81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117

Crushed Dolomite and Dolomitic Limestone Crushed Dolomite and Limestone Siliceous and Limestone Gravel Crushed Limestone Limestone Granite Granite Gneiss Granite Gravel Crushed & Uncrushed Siliceous & Calcareous Gravel Crushed Quartzite Crushed Dolomitic & Calcareous Sandstone Dolomite Red Shale/ Claystone Shell fragment Limestone Dolomite/ Granite Gravel Limestone/ Dolomite/ Chert Gravel Crushed Dolomite Crushed Basalt Granite Crushed Granite Crushed Biotite Granite Gneiss Crushed Fossil Limestone Limestone and Siliceous Gravel Crushed Quartzite Crushed Siliceous Gravel Quartzite (Quartzo Sandstone) Traprock "Limestone Rock Asphalt" Crushed Asphaltic Limestone (Precoated) Limestone Crushed Limestone and Dolomite Crushed Quartzite Gravel Crushed Phyllite (Schist) Crushed Granite Crushed Siliceous Gravel (Granite & Quartzite) Crushed Limestone and Dolomite (with trace of chert) Siliceous gravel Sandstone

308

Appendix L:

Performance Data

309

HMA 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 POOR POOR GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD FAIR FAIR GOOD GOOD FAIR POOR GOOD POOR FAIR POOR GOOD FAIR FAIR

PCC

BC GOOD

OGFC FAIR GOOD FAIR

GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD

GOOD GOOD GOOD GOOD

GOOD GOOD

GOOD

GOOD GOOD GOOD GOOD

GOOD GOOD GOOD GOOD FAIR*

GOOD GOOD GOOD

GOOD

GOOD POOR

310

33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68

POOR* FAIR GOOD POOR* GOOD FAIR GOOD POOR POOR GOOD FAIR FAIR FAIR FAIR FAIR FAIR FAIR GOOD FAIR GOOD FAIR* FAIR* GOOD GOOD GOOD POOR FAIR GOOD POOR GOOD GOOD POOR* GOOD* GOOD GOOD GOOD GOOD GOOD GOOD GOOD

GOOD GOOD GOOD GOOD GOOD

GOOD

GOOD GOOD GOOD GOOD GOOD

GOOD GOOD GOOD GOOD GOOD POOR

311

69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104

GOOD GOOD GOOD FAIR GOOD GOOD FAIR POOR GOOD GOOD GOOD FAIR GOOD GOOD GOOD GOOD FAIR GOOD GOOD GOOD POOR FAIR* GOOD GOOD POOR POOR GOOD POOR POOR FAIR GOOD GOOD FAIR* POOR FAIR GOOD GOOD GOOD GOOD GOOD GOOD GOOD POOR POOR GOOD POOR GOOD GOOD GOOD POOR GOOD GOOD GOOD GOOD FAIR POOR

GOOD GOOD GOOD

GOOD GOOD GOOD

GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD FAIR* GOOD GOOD GOOD POOR FAIR* GOOD GOOD GOOD POOR GOOD GOOD

GOOD GOOD

312

105 106 107 108 109 110 111 112 113 114 115 116 117

GOOD POOR POOR GOOD* GOOD GOOD GOOD FAIR GOOD GOOD GOOD

GOOD

POOR POOR GOOD GOOD POOR GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD GOOD

313

Appendix M:

Hot-Mix Asphalt Aggregates Graphs

315

40.0 35.0 30.0 Micro-Deval, % Loss 25.0 20.0 15.0 10.0 5.0 0.0
Poor Fair Perform ance Good

Figure M.1: Micro-Deval vs. Performance

80.0 Magnesium Sulfate Soundness, % Loss 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0
Poor

Perform ance Fair

Good

Figure M.2: Magnesium Sulfate Soundness vs. Performance

316

40

35 LA Abrasion, % Loss

30

25 20

15

10

Poor

Fair Perform ance

Good

Figure M.3: L.A. Abrasion vs. Performance

16.0 14.0 Canadian Freeze-Thaw, % Loss 12.0 10.0 8.0 6.0 4.0 2.0 0.0
Poor Fair Perform ance Good

Figure M.4: Canadian Freeze-Thaw Soundness vs. Performance

317

35 Aggregate Crushing Value, % Loss

30

25

20

15

10

Poor

Fair Perform ance

Good

Figure M.5: Aggregate Crushing Value vs. Performance

60 Aggregate Crushing Value (SSD), % Loss

50 40

30

20 10

Poor

Fair Perform ance

Good

Figure M.6: Aggregate Crushing Value (SSD) vs. Performance

318

6.0

5.0 4.0 Absorption

3.0

2.0 1.0

0.0

Poor

Fair Perform ance

Good

Figure M.7: Absorption vs. Performance

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 2.10
Poor Fair Perform ance Good

Figure M.8: Specific Gravity (Bulk) vs. Performance

319

3.20 3.10 3.00 Specific Gravity (SSD) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20
Poor Fair Perform ance Good

Figure M.9: Specific Gravity (SSD) vs. Performance

3.10 3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30
Poor Fair Perform ance Good

Figure M.10: Specific Gravity (Apparent) vs. Performance

320

5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00
Poor Fair Perform ance Good

Figure M.11: Particle Shape Factor vs. Performance

120

100

Percent Crushed (1+)

80

60

40

20

Poor

Fair Perform ance

Good

Figure M.12: Particle Shape Factor vs. Performance

321

120

100

Percent Crushed (2+)

80

60

40

20

Poor

Fair Perform ance

Good

Figure M.13: Particle Shape Factor vs. Performance

POOR % of total 100 90 - 99 80 - 89 70 - 79 60 - 69 50 - 59 40 - 49 30 - 39 20 - 29 10 - 19 0 - 10 number 14 2 0 4 0 0 0 0 0 0 0 overall % 70% 10% 0% 20% 0% 0% 0% 0% 0% 0% 0%

Fractured Particles (1 or more sides) FAIR GOOD number 18 3 0 2 1 0 0 1 0 0 1 overall % 69% 12% 0% 8% 4% 0% 0% 4% 0% 0% 4% number 42 5 1 0 2 0 0 0 0 1 1 overall % 81% 10% 2% 0% 4% 0% 0% 0% 0% 2% 2% POOR 19% 20% 0% 67% 0% 0% 0% 0% 0% 0% 0%

Overall % FAIR 24% 30% 0% 33% 33% 0% 0% 100% 0% 0% 50% GOOD 57% 50% 100% 0% 67% 0% 0% 0% 0% 100% 50%

Table M.1:

Percent Fractured Particles (1 or more sides)

322

POOR % of total 100 90 - 99 80 - 89 70 - 79 60 - 69 50 - 59 40 - 49 30 - 39 20 - 29 10 - 19 0 - 10 number 14 2 0 3 1 0 0 0 0 0 0 overall % 70% 10% 0% 15% 5% 0% 0% 0% 0% 0% 0%

Fractured Particles (2 or more sides) FAIR GOOD number 17 3 1 1 1 1 0 0 1 0 1 overall % 65% 12% 4% 4% 4% 4% 0% 0% 4% 0% 4% number 42 4 1 1 2 0 0 0 0 0 2 overall % 81% 8% 2% 2% 4% 0% 0% 0% 0% 0% 4% POOR 19% 22% 0% 60% 25% 0% 0% 0% 0% 0% 0%

Overall % FAIR 23% 33% 50% 20% 25% 100% 0% 0% 100% 0% 33% GOOD 58% 44% 50% 20% 50% 0% 0% 0% 0% 0% 67%

Table M.2:

Percent Fractured Particles (2 or more sides)

323

40.0 Magnesium Sulfate Soundness, % Loss

POOR
35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

FAIR GOOD

Figure M.14: Magnesium Sulfate Soundness vs. Micro-Deval

70 60 LA Abrasion, % Loss 50 40 30 20 10 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure M.15: L.A. Abrasion vs. Micro-Deval

324

16.0

POOR
14.0 Canadian Freeze-Thaw, % Loss 12.0 10.0 8.0 6.0 4.0 2.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

FAIR GOOD

Figure M.16: Canadian Freeze-Thaw Soundness vs. Micro-Deval

50 Aggregate Crushing Value, % Loss 45 40 35 30 25 20 15 10 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure M.17: Aggregate Crushing Value vs. Micro-Deval

325

55 Aggregate Crushing Value (SSD), % Loss 50 45 40 35 30 25 20 15 10 5 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure M.18: Aggregate Crushing Value (SSD) vs. Micro-Deval

6.0

5.0

4.0 Absorption

3.0

2.0

POOR
1.0

FAIR GOOD

0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

Figure M.19: Absorption vs. Micro-Deval

326

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure M.20: Specific Gravity (Bulk) vs. Micro-Deval

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure M.21: Specific Gravity (SSD) vs. Micro-Deval

327

3.10

POOR
3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

FAIR GOOD

Figure M.22: Specific Gravity (Apparent) vs. Micro-Deval

5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure M.23: Particle Shape Factor vs. Micro-Deval

328

120

100 Percent Crushed (1+)

80

60

40

POOR
20

FAIR GOOD

0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

Figure M.24: Percent Crushed (1+ Sides) vs. Micro-Deval

120

100 Percent Crushed (2+)

80

60

40

POOR
20

FAIR GOOD

0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

Figure M.25: Percent Crushed (2+ Sides) vs. Micro-Deval

329

70 60 LA Abrasion, % Loss 50 40 30 20 10 0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure M.26: L.A. Abrasion vs. Magnesium Sulfate Soundness

16.0 14.0 Canadian Freeze-Thaw, % Loss 12.0 10.0 8.0 6.0 4.0 2.0 0.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure M.27: Canadian Freeze-Thaw Soundness vs. Magnesium Sulfate Soundness

330

60

POOR
Aggregate Crushing Value, % Loss 50 40 30 20 10 0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

FAIR GOOD

Figure M.28: Aggregate Crushing Value vs. Magnesium Sulfate Soundness

60 Aggregate Crushing Value (SSD), % Loss 50 40

30 20

POOR
10 0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

FAIR GOOD

Figure M.29: Aggregate Crushing Value (SSD) vs. Magnesium Sulfate Soundness

331

6.0 5.0

4.0 Absorption 3.0 2.0

POOR
1.0 0.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

FAIR GOOD

Figure M.30: Absorption vs. Magnesium Sulfate Soundness

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure M.31: Specific Gravity (Bulk) vs. Magnesium Sulfate Soundness

332

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure M.32: Specific Gravity (SSD) vs. Magnesium Sulfate Soundness

3.10

POOR
3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

FAIR GOOD

Figure M.33: Specific Gravity (Apparent) vs. Magnesium Sulfate Soundness

333

5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure M.34: Particle Shape Factor vs. Magnesium Sulfate Soundness

16.0

POOR
14.0 Canadian Freeze-Thaw, % Loss 12.0 10.0 8.0 6.0 4.0 2.0 0.0 10 20 30 40 LA Abrasion, % Loss 50 60 70

FAIR GOOD

Figure M.35: Canadian Freeze-Thaw Soundness vs. L.A. Abrasion

334

50 Aggregate Crushing Value, % Loss 45 40 35 30 25 20 15 10 10 20 30 40 LA Abrasion, % Loss 50 60 70

POOR FAIR GOOD

Figure M.36: Aggregate Crushing Value vs. L.A. Abrasion

55 Aggregate Crushing Value (SSD), % Loss 50 45 40 35 30 25 20 15 10 10 20 30 40 LA Abrasion, % Loss 50 60 70

POOR FAIR GOOD

Figure M.37: Aggregate Crushing Value (SSD) vs. L.A. Abrasion

335

6.0 5.0 4.0 Absorption

3.0 2.0

POOR
1.0 0.0 10 20 30 40 LA Abrasion, % Loss 50 60 70

FAIR GOOD

Figure M.38: Absorption vs. L.A. Abrasion

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 10 20 30 40 LA Abrasion, % Loss 50 60 70

POOR FAIR GOOD

Figure M.39: Specific Gravity (Bulk) vs. L.A. Abrasion

336

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 10 20 30 40 LA Abrasion, % Loss 50 60 70

POOR FAIR GOOD

Figure M.40: Specific Gravity (SSD) vs. L.A. Abrasion

3.10 3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 10 20 30 40 LA Abrasion, % Loss 50 60 70

POOR FAIR GOOD

Figure M.41: Specific Gravity (Apparent) vs. L.A. Abrasion

337

5.50 5.00 4.50 Particle Shaper Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 10 20 30 40 LA Abrasion, % Loss 50 60 70

POOR FAIR GOOD

Figure M.42: Particle Shape Factor vs. L.A. Abrasion

50

POOR
Aggregate Crushing Value, % Loss 45 40 35 30 25 20 15 10 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 Canadian Freeze-Thaw , % Loss

FAIR GOOD

Figure M.43: Aggregate Crushing Value vs. Canadian Freeze-Thaw Soundness

338

60 Aggregate Crushing Value (SSD), % Loss

POOR
50 40 30 20 10 0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 Canadian Freeze-Thaw , % Loss

FAIR GOOD

Figure M.44: Aggregate Crushing Value (SSD) vs. Canadian Freeze-Thaw Soundness

6.0

POOR
5.0 4.0 Absorption 3.0 2.0 1.0 0.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 Canadian Freeze-Thaw , % Loss

FAIR GOOD

Figure M.45: Absorption vs. Canadian Freeze-Thaw Soundness

339

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 Canadian Freeze-Thaw , % Loss

POOR FAIR GOOD

Figure M.46: Specific Gravity (Bulk) vs. Canadian Freeze-Thaw Soundness

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 Canadian Freeze-Thaw , % Loss

POOR FAIR GOOD

Figure M.47: Specific Gravity (SSD) vs. Canadian Freeze-Thaw Soundness

340

3.10 3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 Canadian Freeze-Thaw , % Loss

POOR FAIR GOOD

Figure M.48: Specific Gravity (Apparent) vs. Canadian Freeze-Thaw Soundness

5.50 5.00 4.50 Percent Particle Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 1.0 3.0 5.0 7.0 9.0 11.0 13.0 15.0 Canadian Freeze-Thaw , % Loss

POOR FAIR GOOD

Figure M.49: Particle Shape Factor vs. Canadian Freeze-Thaw Soundness

341

60 Aggregate Crushing Value (SSD), % Loss 50 40 30 20

POOR
10 0 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

FAIR GOOD

Figure M.50: Aggregate Crushing Value (SSD) vs. Aggregate Crushing Value

6.0 5.0 4.0 Absorption 3.0 2.0

POOR
1.0 0.0 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

FAIR GOOD

Figure M.51: Absorption vs. Aggregate Crushing Value

342

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

Figure M.52: Specific Gravity (Bulk) vs. Aggregate Crushing Value

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

Figure M.53: Specific Gravity (SSD) vs. Aggregate Crushing Value

343

3.10

POOR
3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

FAIR GOOD

Figure M.54: Specific Gravity (Apparent) vs. Aggregate Crushing Value

5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

Figure M.55: Particle Shape Factor vs. Aggregate Crushing Value

344

6.0 5.0 4.0 Absorption 3.0 2.0

POOR
1.0 0.0 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

FAIR GOOD

Figure M.56: Absorption vs. Aggregate Crushing Value (SSD)

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

Figure M.57: Specific Gravity (Bulk) vs. Aggregate Crushing Value (SSD)

345

3.10 3.00 Specific Gravity (SSD) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 5 10 15 20 25 30 35 40 45

POOR FAIR GOOD

50

Aggregate Crushing Value (SSD), % Loss

Figure M.58: Specific Gravity (SSD) vs. Aggregate Crushing Value (SSD)

3.10 3.00 Specific Gravity (Apparent) 2.90

POOR FAIR GOOD

2.80 2.70 2.60 2.50 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

Figure M.59: Specific Gravity (Apparent) vs. Aggregate Crushing Value (SSD)

346

5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

Figure M.60: Particle Shape Factor vs. Aggregate Crushing Value (SSD)

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 1.0 2.0 3.0 Absorption 4.0 5.0 6.0

POOR FAIR GOOD

Figure M.61: Specific Gravity (Bulk) vs. Absorption

347

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 1.0 2.0 3.0 Absorption 4.0 5.0 6.0

POOR FAIR GOOD

Figure M.62: Specific Gravity (SSD) vs. Absorption

3.10

POOR
3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 0.0 1.0 2.0 3.0 Absorption 4.0 5.0 6.0

FAIR GOOD

Figure M.63: Specific Gravity (Apparent) vs. Absorption

348

5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.0 1.0 2.0 3.0 Absorption 4.0 5.0 6.0

POOR FAIR GOOD

Figure M.64: Particle Shape Factor vs. Absorption

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10

Specific Gravity (Bulk)

Figure M.65: Specific Gravity (SSD) vs. Specific Gravity (Bulk)

349

3.10 3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10

Specific Gravity (Bulk)

Figure M.66: Specific Gravity (Apparent) vs. Specific Gravity (Bulk)

5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 2.20

POOR FAIR GOOD

2.30

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (Bulk)

Figure M.67: Particle Shape Factor vs. Specific Gravity (Bulk)

350

3.10 3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10

Specific Gravity (SSD)

Figure M.68: Specific Gravity (Apparent) vs. Specific Gravity (SSD)

5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 2.20

POOR FAIR GOOD

2.30

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (SSD)

Figure M.69: Particle Shape Factor vs. Specific Gravity (SSD)

351

5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 2.30

POOR FAIR GOOD

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (Apparent)

Figure M.70: Particle Shape Factor vs. Specific Gravity (Apparent)

352

Appendix N:

Portland Cement Concrete Aggregates Graphs

353

40.0 35.0 30.0 Micro-Deval, % Loss 25.0 20.0 15.0 10.0 5.0 0.0

Poor

Fair
Performance

Good

Figure N.1: Micro-Deval vs. Performance

35.0 Magnesium Sulfate Soundness, % Loss 30.0 25.0 20.0 15.0 10.0 5.0 0.0

Poor

Fair
Performance

Good

Figure N.2: Magnesium Sulfate Soundness vs. Performance

354

70 60 50 40 30 20 10 0

LA Abrasion, % Loss

Poor

Fair Performance

Good

Figure N.3: L.A. Abrasion vs. Performance

14.0 12.0 Canadian Freeze-Thaw, % Loss 10.0 8.0 6.0 4.0 2.0 0.0

Poor

Fair
Performance

Good

Figure N.4: Canadian Freeze-Thaw vs. Performance

355

50 45 Aggregate Crushing Value, % Loss 40 35 30 25 20 15 10 5 0

Poor

Fair
Performance

Good

Figure N.5: Aggregate Crushing Value vs. Performance

60 Aggregate Crushing Value (SSD), % Loss

50

40

30

20

10

Poor

Fair
Performance

Good

Figure N.6: Aggregate Crushing Value (SSD) vs. Performance

356

5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0

Poor

Fair
Performance

Good

Figure N.7: Absorption vs. Performance

2.90 2.80 2.70

Specific Gravity (Bulk)

2.60 2.50

2.40 2.30 2.20

Poor

Fair
Performance

Good

Figure N.8: Specific Gravity (Bulk) vs. Performance

357

3.00

2.90 2.80 2.70 2.60

Specific Gravity (SSD)

2.50 2.40 2.30

Poor

Fair
Performance

Good

Figure N.9: Specific Gravity (SSD) vs. Performance

3.00

2.90 Specific Gravity (Apparent)

2.80

2.70

2.60

Poor
2.50

Fair

Good

Performance

Figure N.10: Specific Gravity (Apparent) vs. Performance

358

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00

Poor

Fair
Performance

Good

Figure N.11: Particle Shape Factor vs. Performance

POOR % of total 100 90 - 99 80 - 89 70 - 79 60 - 69 50 - 59 40 - 49 30 - 39 20 - 29 10 - 19 0 - 10 number 2 1 0 2 0 0 0 1 0 0 0 overall % 33% 17% 0% 33% 0% 0% 0% 17% 0% 0% 0%

Fractured Particles (1 or more sides) FAIR GOOD number 3 0 0 0 0 0 0 0 0 1 0 overall % 75% 0% 0% 0% 0% 0% 0% 0% 0% 25% 0% number 29 2 1 1 2 0 1 0 0 1 2 overall % 74% 5% 3% 3% 5% 0% 3% 0% 0% 3% 5% POOR 6% 33% 0% 67% 0% 0% 0% 100% 0% 0% 0%

Overall % FAIR 9% 0% 0% 0% 0% 0% 0% 0% 0% 50% 0% GOOD 85% 67% 100% 33% 100% 0% 100% 0% 0% 50% 100%

Table M.1: Fractured Particles (1 or more sides) vs. Performance

359

POOR % of total 100 90 - 99 80 - 89 70 - 79 60 - 69 50 - 59 40 - 49 30 - 39 20 - 29 10 - 19 0 - 10 number 2 0 1 2 0 0 0 0 1 0 0 overall % 33% 0% 17% 33% 0% 0% 0% 0% 17% 0% 0%

Fractured Particles (2 or more sides) FAIR GOOD number 3 0 0 0 0 0 0 0 0 0 1 overall % 75% 0% 0% 0% 0% 0% 0% 0% 0% 0% 25% number 29 2 0 2 2 0 0 1 0 0 3 overall % 74% 5% 0% 5% 5% 0% 0% 3% 0% 0% 8% POOR 6% 0% 100% 50% 0% 0% 0% 0% 100% 0% 0%

Overall % FAIR 9% 0% 0% 0% 0% 0% 0% 0% 0% 0% 25% GOOD 85% 100% 0% 50% 100% 0% 0% 100% 0% 0% 75%

Table M.2: Fractured Particles (2 or more sides) vs. Performance

35.0 Magnesium Sulfate Soundness, % Loss 30.0 25.0 20.0 15.0 10.0 5.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure N.12: Magnesium Sulfate Soundness vs. Micro-Deval

360

70 60 50 40 30 20 10 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

LA Abrasion, % Loss

POOR FAIR GOOD

40.0

Figure N.13: L.A. Abrasion vs. Micro-Deval

14.0 12.0 Canadian Freeze-Thaw, % Loss 10.0 8.0 6.0 4.0 2.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure N.14: Canadian Freeze-Thaw vs. Micro-Deval

361

50 45 Aggregate Crushing Value, % Loss 40 35 30 25 20 15 10 5 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure N.15: Aggregate Crushing Value vs. Micro-Deval

60 Aggregate Crushing Value (SSD), % Loss

50

40

30

20

10

POOR FAIR GOOD

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0

0 Micro-Deval, % Loss

Figure N.16: Aggregate Crushing Value (SSD) vs. Micro-Deval

362

5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure N.17: Absorption vs. Micro-Deval

3.00 2.90 2.80 Specific Gravity (Bulk) 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure N.18: Specific Gravity (Bulk) vs. Micro-Deval

363

3.00 2.90 2.80 Specific Gravity (SSD) 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure N.19: Specific Gravity (SSD) vs. Micro-Deval

3.00 2.90 Specific Gravity (Apparent) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure N.20: Specific Gravity (Apparent) vs. Micro-Deval 364

4.00

3.50

Particle Shape Factor

3.00

2.50

2.00

1.50

1.00 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure N.21: Particle Shape Factor vs. Micro-Deval

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure N.22: Fractured Particles (1 or more sides) vs. Micro-Deval 365

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Micro-Deval, % Loss

POOR FAIR GOOD

40.0

Figure N.23: Fractured Particles (2 or more sides) vs. Micro-Deval

70 60 50 40 30 20 10 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Magnesium Sulfate Soundness, % Loss

LA Abrasion, % Loss

POOR FAIR GOOD

Figure N.24: L.A. Abrasion vs. Magnesium Sulfate Soundness

366

14.0 12.0 Canadian Freeze-Thaw, % Loss 10.0 8.0 6.0 4.0 2.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure N.25: Canadian Freeze-Thaw vs. Magnesium Sulfate Soundness

50 45 Aggregate Crushing Value, % Loss 40 35 30 25 20 15 10 5 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure N.26: Aggregate Crushing Value vs. Magnesium Sulfate Soundness

367

60 Aggregate Crushing Value (SSD), % Loss

50

40

30

20

10

POOR FAIR GOOD

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0

0 Magnesium Sulfate Soundness, % Loss

Figure N.27: Aggregate Crushing Value (SSD) vs. Magnesium Sulfate Soundness

5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

35.0

Figure N.28: Absorption vs. Magnesium Sulfate Soundness

368

3.00 2.90 2.80 Specific Gravity (Bulk) 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure N.29: Specific Gravity (Bulk) vs. Magnesium Sulfate Soundness

3.00 2.90 2.80 Specific Gravity (SSD) 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure N.30: Specific Gravity (SSD) vs. Magnesium Sulfate Soundness 369

3.00 2.90 Specific Gravity (Apparent) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure N.31: Specific Gravity (Apparent) vs. Magnesium Sulfate Soundness

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure N.32: Particle Shape Factor vs. Magnesium Sulfate Soundness

370

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure N.33: Fractured Particles (1 or more sides) vs. Magnesium Sulfate Soundness

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure N.34: Fractured Particles (2 or more sides) vs. Magnesium Sulfate Soundness 371

14.0 12.0 Canadian Freeze-Thaw, % Loss 10.0 8.0 6.0 4.0 2.0 0.0 0 10 20 30 40 50 60 LA Abrasion, % Loss

POOR FAIR GOOD

70

Figure N.35: Canadian Freeze-Thaw vs. L.A. Abrasion

50 45 Aggregate Crushing Value, % Loss 40 35 30 25 20 15 10 5 0 0 10 20 30 40 50 60 70 LA Abrasion, % Loss

POOR FAIR GOOD

Figure N.36: Aggregate Crushing Value vs. L.A. Abrasion

372

60 Aggregate Crushing Value (SSD), % Loss

50

40

30

20

10

POOR FAIR GOOD

0 10 20 30 40 50 60 70

0 LA Abrasion, % Loss

Figure N.37: Aggregate Crushing Value (SSD) vs. L.A. Abrasion

5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0 10 20 30 40 50 60 70 LA Abrasion, % Loss

POOR FAIR GOOD

Figure N.38: Absorption vs. L.A. Abrasion 373

3.00 2.90 2.80 Specific Gravity (Bulk) 2.70 2.60 2.50 2.40 2.30 2.20 0 10 20 30 40 50 60 LA Abrasion, % Loss

POOR FAIR GOOD

70

Figure N.39: Specific Gravity (Bulk) vs. L.A. Abrasion

3.00 2.90 2.80 Specific Gravity (SSD) 2.70 2.60 2.50 2.40 2.30 2.20 0 10 20 30 40 50 60 LA Abrasion, % Loss

POOR FAIR GOOD

70

Figure N.40: Specific Gravity (SSD) vs. L.A. Abrasion

374

3.00 2.90 Specific Gravity (Apparent) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0 10 20 30 40 50 60 LA Abrasion, % Loss

POOR FAIR GOOD

70

Figure N.41: Specific Gravity (Apparent) vs. L.A. Abrasion

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0 10 20 30 40 50 60 LA Abrasion, % Loss

POOR FAIR GOOD

70

Figure N.42: Particle Shape Factor vs. L.A. Abrasion

375

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 LA Abrasion, % Loss

POOR FAIR GOOD

Figure N.43: Fractured Particles (1 or more sides) vs. L.A. Abrasion

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 LA Abrasion, % Loss

POOR FAIR GOOD

Figure N.44: Fractured Particles (2 or more sides) vs. L.A. Abrasion

376

50 45 Aggregate Crushing Value, % Loss 40 35 30 25 20 15 10 5 0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

14.0

Figure N.45: Aggregate Crushing Value vs. Canadian Freeze-Thaw

60 Aggregate Crushing Value (SSD), % Loss

50

40

30

20

10

POOR FAIR GOOD

0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0

0 Canadian Freeze-Thaw, % Loss

Figure N.46: Aggregate Crushing Value (SSD) vs. Canadian Freeze-Thaw

377

5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

Figure N.47: Absorption vs. Canadian Freeze-Thaw

3.00 2.90 2.80 Specific Gravity (Bulk) 2.70 2.60 2.50 2.40 2.30 2.20 0.0 2.0 4.0 6.0 8.0 10.0 12.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

14.0

Figure N.48: Specific Gravity (Bulk) vs. Canadian Freeze-Thaw

378

3.00 2.90 2.80 Specific Gravity (SSD) 2.70 2.60 2.50 2.40 2.30 2.20 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

Figure N.49: Specific Gravity (SSD) vs. Canadian Freeze-Thaw

3.00 2.90 Specific Gravity (Apparent) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

Figure N.50: Specific Gravity (Apparent) vs. Canadian Freeze-Thaw

379

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

Figure N.51: Particle Shape Factor vs. Canadian Freeze-Thaw

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

14.0

Figure N.52: Fractured Particles (1 or more sides) vs. Canadian Freeze-Thaw

380

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

Figure N.53: Fractured Particles (2 or more sides) vs. Canadian Freeze-Thaw

60 Aggregate Crushing Value (SSD), % Loss

50

40

30

20

10

POOR FAIR GOOD

0 5 10 15 20 25 30 35 40 45 50

0 Aggregate Crushing Value, % Loss

Figure N.54: Aggregate Crushing Value (SSD) vs. Aggregate Crushing Value

381

5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

Figure N.55: Absorption vs. Aggregate Crushing Value

3.00 2.90 2.80 Specific Gravity (Bulk) 2.70 2.60 2.50 2.40 2.30 2.20 0 5 10 15 20 25 30 35 40 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

45 50

Figure N.56: Specific Gravity (Bulk) vs. Aggregate Crushing Value

382

3.00 2.90 2.80 Specific Gravity (SSD) 2.70 2.60 2.50 2.40 2.30 2.20 0 5 10 15 20 25 30 35 40 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

45 50

Figure N.57: Specific Gravity (SSD) vs. Aggregate Crushing Value

3.00 2.90 Specific Gravity (Apparent) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0 5 10 15 20 25 30 35 40 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

45 50

Figure N.58: Specific Gravity (Apparent) vs. Aggregate Crushing Value

383

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0 5 10 15 20 25 30 35 40 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

45 50

Figure N.59: Particle Shape Factor vs. Aggregate Crushing Value

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 5 10 15 20 25 30 35 40 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

45 50

Figure N.60: Fractured Particles (1 or more sides) vs. Aggregate Crushing Value

384

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 5 10 15 20 25 30 35 40 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

45 50

Figure N.61: Fractured Particles (2 or more sides) vs. Aggregate Crushing Value

5.0 4.5 4.0 3.5 Absorption, % 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0 10 20 30 40 50 60 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

Figure N.62: Absorption vs. Aggregate Crushing Value (SSD)

385

3.00 2.90 2.80 Specific Gravity (Bulk) 2.70 2.60 2.50 2.40 2.30 2.20 0 10 20 30 40 50 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

60

Figure N.63: Specific Gravity (Bulk) vs. Aggregate Crushing Value (SSD)

3.00 2.90 2.80 Specific Gravity (SSD) 2.70 2.60 2.50 2.40 2.30 2.20 0 10 20 30 40 50 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

60

Figure N.64: Specific Gravity (SSD) vs. Aggregate Crushing Value (SSD)

386

3.00 2.90 Specific Gravity (Apparent) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0 10 20 30 40 50 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

60

Figure N.65: Specific Gravity (Apparent) vs. Aggregate Crushing Value (SSD)

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0 10 20 30 40 50 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

60

Figure N.66: Particle Shape Factor vs. Aggregate Crushing Value (SSD)

387

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

60

Figure N.67: Fractured Particles (1 or more sides) vs. Aggregate Crushing Value (SSD)
100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

60

Figure N.68: Fractured Particles (2 or more sides) vs. Aggregate Crushing Value (SSD)

388

3.00 2.90 2.80 Specific Gravity (Bulk) 2.70 2.60 2.50 2.40 2.30 2.20 0.0 0.5 1.0 1.5 2.0 2.5 Absorption, % 3.0 3.5 4.0 4.5 5.0

POOR FAIR GOOD

Figure N.69: Specific Gravity (Bulk) vs. Absorption

3.00 2.90 2.80 Specific Gravity (SSD) 2.70 2.60 2.50 2.40 2.30 2.20 0.0 0.5 1.0 1.5 2.0 2.5 Absorption, % 3.0 3.5 4.0 4.5 5.0

POOR FAIR GOOD

Figure N.70: Specific Gravity (SSD) vs. Absorption 389

3.00 2.90 Specific Gravity (Apparent) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 0.5 1.0 1.5 2.0 2.5 Absorption, % 3.0 3.5 4.0 4.5 5.0

POOR FAIR GOOD

Figure N.71: Specific Gravity (Apparent) vs. Absorption

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0.0 0.5 1.0 1.5 2.0 2.5 Absorption, % 3.0 3.5 4.0 4.5 5.0

POOR FAIR GOOD

Figure N.72: Particle Shape Factor vs. Absorption 390

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 0.5 1.0 1.5 2.0 2.5 Absorption, % 3.0 3.5 4.0 4.5 5.0

POOR FAIR GOOD

Figure N.73: Fractured Particles (1 or more sides) vs. Absorption

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 0.5 1.0 1.5 2.0 2.5 Absorption, % 3.0 3.5 4.0 4.5 5.0

POOR FAIR GOOD

Figure N.74: Fractured Particles (2 or more sides) vs. Absorption 391

3.00 2.90 2.80 Specific Gravity (SSD) 2.70 2.60 2.50 2.40 2.30 2.20 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00

Specific Gravity (Bulk)

Figure N.75: Specific Gravity (SSD) vs. Specific Gravity (Bulk)

3.00 2.90 Specific Gravity (Apparent) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00

Specific Gravity (Bulk)

Figure N.76: Specific Gravity (Apparent) vs. Specific Gravity (Bulk)

392

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00

Specific Gravity (Bulk)

Figure N.77: Particle Shape Factor vs. Specific Gravity (Bulk)

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00

Specific Gravity (Bulk)

Figure N.78: Fractured Particles (1 or more sides) vs. Specific Gravity (Bulk)

393

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00

Specific Gravity (Bulk)

Figure N.79: Fractured Particles (2 or more sides) vs. Specific Gravity (Bulk)

3.00 2.90 Specific Gravity (Apparent) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00

Specific Gravity (SSD)

Figure N.80: Specific Gravity (Apparent) vs. Specific Gravity (SSD)

394

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 Specific Gravity (SSD)

Figure N.81: Particle Shape Factor vs. Specific Gravity (SSD)

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 Specific Gravity (SSD)

Figure N.82: Fractured Particles (1 or more sides) vs. Specific Gravity (SSD)

395

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 Specific Gravity (SSD)

Figure N.83: Fractured Particles (2 or more sides) vs. Specific Gravity (SSD)

4.50 4.00 3.50 Particle Shape Factor 3.00 2.50 2.00 1.50 1.00 0.50 0.00 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00

Specific Gravity (Apparent)

Figure N.84: Particle Shape Factor vs. Specific Gravity (Apparent)

396

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00

Specific Gravity (Apparent)

Figure N.85: Fractured Particles (1 or more sides) vs. Specific Gravity (Apparent)

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 Specific Gravity (Apparent)

Figure N.86: Fractured Particles (2 or more sides) vs. Specific Gravity (Apparent)

397

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 1.00

POOR FAIR GOOD

1.50 2.00 2.50 Particle Shape Factor 3.00 3.50 4.00

Figure N.87: Fractured Particles (1 or more sides) vs. Particle Shape Factor

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 1.00

POOR FAIR GOOD

1.50 2.00 2.50 Particle Shape Factor 3.00 3.50 4.00

Figure N.88: Fractured Particles (2 or more sides) vs. Particle Shape Factor

398

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 Fractured Particles (1 or more sides), % of total

POOR FAIR GOOD

90 100

Figure N.89: Fractured Particles (2 or more sides) vs. Fractured Particles (1 or more sides)

399

Appendix O:

Base Course Aggregates Graphs

401

50.0 45.0 40.0 Micro-Deval, % Loss 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0
Poor Fair Perform ance Good

Figure O.1:
80.0 Magnesium Sulfate Soundness, % Loss 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0
Poor

Micro-Deval vs. Performance

Perform Fair ance

Good

Figure O.2:

Magnesium Sulfate Soundness vs. Performance

402

70

60 LA Abrasion, % Loss

50

40

30

20

10

Poor

Fair Perform ance

Good

Figure O.3:
25.0

L.A. Abrasion vs. Performance

Canadian Freeze-Thaw, % Loss

20.0

15.0

10.0

5.0

0.0
Poor

Perform ance Fair

Good

Figure O.4:

Canadian Freeze-Thaw Soundness vs. Performance

403

35 Aggregate Crushing Value, % Loss

30

25

20

15

10

Poor

Fair Perform ance

Good

Figure O.5:
45 Aggregate Crushing Value (SSD), % Loss 40 35 30 25 20 15 10

Aggregate Crushing Value vs. Performance

Poor

Perform Fair ance

Good

Figure O.6:

Aggregate Crushing Value (SSD) vs. Performance

404

6.0

5.0

4.0 Absorption

3.0

2.0

1.0

0.0

Poor

Fair Perform ance

Good

Figure O.7:

Absorption vs. Performance

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20
Poor Fair Perform ance Good

Figure O.8:

Specific Gravity (Bulk) vs. Performance

405

3.10 3.00 Specific Gravity (SSD) 2.90 2.80 2.70 2.60 2.50 2.40 2.30
Poor Fair Perform ance Good

Figure O.9:

Specific Gravity (SSD) vs. Performance

3.10

3.00 Specific Gravity (Apparent)

2.90

2.80 2.70

2.60

2.50

Poor

Fair Perform ance

Good

Figure O.10: Specific Gravity (Apparent) vs. Performance

406

4.50

4.00 Particle Shape Factor 3.50

3.00

2.50 2.00

1.50

Poor

Fair Perform ance

Good

Figure O.11: Particle Shape Factor vs. Performance

120

100

% Crushed (1+)

80

60 40

20

Poor

Fair Perform ance

Good

Figure O.12: Particle Shape Factor vs. Performance

407

120

100

% Crushed (1+)

80

60

40

20

Poor

Fair Perform ance

Good

Figure O.13: Particle Shape Factor vs. Performance

Fractured Particles (1 or more sides) FAIR GOOD number 1 0 0 1 0 0 0 0 0 0 0 overall % 50% 0% 0% 50% 0% 0% 0% 0% 0% 0% 0% number 44 5 1 3 3 0 0 1 0 1 1 overall % 75% 8% 2% 5% 5% 0% 0% 2% 0% 2% 2% POOR 8% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%

POOR % of total 100 90 - 99 80 - 89 70 - 79 60 - 69 50 - 59 40 - 49 30 - 39 20 - 29 10 - 19 0 - 10 number 4 0 0 0 0 0 0 0 0 0 0 overall % 100% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Overall % FAIR 2% 0% 0% 25% 0% 0% 0% 0% 0% 0% 0% GOOD 90% 100% 100% 75% 100% 0% 0% 100% 0% 100% 100%

Table O.1:

Percent Fractured Particles (1 or more sides)

408

POOR % of total 100 90 - 99 80 - 89 70 - 79 60 - 69 50 - 59 40 - 49 30 - 39 20 - 29 10 - 19 0 - 10 number 4 0 0 0 0 0 0 0 0 0 0 overall % 100% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Fractured Particles (2 or more sides) FAIR GOOD number 1 0 0 1 0 0 0 0 0 0 0 overall % 50% 0% 0% 50% 0% 0% 0% 0% 0% 0% 0% number 44 5 0 3 3 1 0 0 1 0 2 overall % 75% 8% 0% 5% 5% 2% 0% 0% 2% 0% 3% POOR 8% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Overall % FAIR 2% 0% 0% 25% 0% 0% 0% 0% 0% 0% 0% GOOD 90% 100% 0% 75% 100% 100% 0% 0% 100% 0% 100%

Table O.2:

Percent Fractured Particles (2 or more sides)

409

80.0 Magnesium Sulfate Soundness, % Loss 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure O.14: Magnesium Sulfate Soundness vs. Micro-Deval

70

POOR
60 LA Abrasion, % Loss 50

FAIR GOOD

40 30

20 10 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

Figure O.15: L.A. Abrasion vs. Micro-Deval

410

25.0

Canadian Freeze-Thaw, % Loss

20.0

15.0

10.0

5.0

POOR FAIR GOOD

0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

Figure O.16: Canadian Freeze-Thaw Soundness vs. Micro-Deval

45 Aggregate Crushing Value, % Loss 40 35 30 25 20 15 10 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure O.17: Aggregate Crushing Value vs. Micro-Deval

411

45 Aggregate Crushing Value (SSD), % Loss 40 35 30 25 20 15 10 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure O.18: Aggregate Crushing Value (SSD) vs. Micro-Deval

6.0

5.0

4.0 Absorption

3.0

2.0

POOR
1.0

FAIR GOOD

0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

Figure O.19: Absorption vs. Micro-Deval

412

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

POOR FAIR GOOD

Figure O.20: Specific Gravity (Bulk) vs. Micro-Deval

3.10

POOR
3.00 Specific Gravity (SSD) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

FAIR GOOD

Figure O.21: Specific Gravity (SSD) vs. Micro-Deval

413

3.10

3.00 Specific Gravity (Apparent)

2.90

2.80

2.70

POOR
2.60

FAIR GOOD

2.50 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Micro-Deval, % Loss

Figure O.22: Specific Gravity (Apparent) vs. Micro-Deval

4.50

POOR
4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Micro-Deval, % Loss

FAIR GOOD

Particle Shape Factor

Figure O.23: Particle Shape Factor vs. Micro-Deval

414

70 60 LA Abrasion, % Loss 50 40 30 20 10 0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure O.24: L.A. Abrasion vs. Magnesium Sulfate Soundness

25.0

Canadian Freeze-Thaw, % Loss

20.0

15.0

10.0

5.0

POOR FAIR GOOD

0.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

Figure O.25: Canadian Freeze-Thaw Soundness vs. Magnesium Sulfate Soundness

415

45 Aggregate Crushing Value, % Loss 40 35 30 25 20 15 10 5 0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

80.0

Figure O.26: Canadian Freeze-Thaw Soundness vs. Magnesium Sulfate Soundness

50 Aggregate Crushing Value (SSD), % Loss 45 40 35 30 25 20 15 10 5 0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

80.0

Figure O.27: Canadian Freeze-Thaw Soundness vs. Magnesium Sulfate Soundness

416

6.0 5.0

4.0 Absorption 3.0 2.0

POOR
1.0 0.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

FAIR GOOD
80.0

Figure O.28: Absorption vs. Magnesium Sulfate Soundness

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure O.29: Specific Gravity (Bulk) vs. Magnesium Sulfate Soundness

417

3.10

POOR
3.00 Specific Gravity (SSD) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

FAIR GOOD

Figure O.30: Specific Gravity (SSD) vs. Magnesium Sulfate Soundness

3.10

POOR
3.00 Specific Gravity (Apparent)

FAIR GOOD

2.90

2.80 2.70

2.60

2.50 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

Figure O.31: Specific Gravity (Apparent) vs. Magnesium Sulfate Soundness

418

4.50 4.00 Particle Shape Factor 3.50 3.00 2.50 2.00 1.50 1.00 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

Figure O.32: Particle Shape Factor vs. Magnesium Sulfate Soundness

25.0

POOR
Canadian Freeze-Thaw, % Loss 20.0

FAIR GOOD

15.0

10.0

5.0

0.0 10 20 30 40 LA Abrasion, % Loss 50 60 70

Figure O.33: Canadian Freeze-Thaw Soundness vs. L.A. Abrasion

419

45 Aggregate Crushing Value, % Loss 40 35 30 25 20 15 10 10 20 30 40 LA Abrasion, % Loss 50 60 70

POOR FAIR GOOD

Figure O.34: Aggregate Crushing Value vs. L.A. Abrasion

45 Aggregate Crushing Value (SSD), % Loss 40 35 30 25 20 15 10 10 20 30 40 LA Abrasion, % Loss 50 60 70

POOR FAIR GOOD

Figure O.35: Aggregate Crushing Value vs. Micro-Deval

420

6.0

POOR
5.0 4.0 Absorption

FAIR GOOD

3.0 2.0 1.0 0.0 10 20 30 40 LA Abrasion, % Loss 50 60 70

Figure O.36: Absorption vs. L.A. Abrasion

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 10 20 30 40 LA Abrasion, % Loss 50 60 70

POOR FAIR GOOD

Figure O.37: Specific Gravity (Bulk) vs. L.A. Abrasion

421

3.10 3.00 Specific Gravity (SSD) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 10 20 30 40 LA Abrasion, % Loss 50 60

POOR FAIR GOOD

70

Figure O.38: Specific Gravity (SSD) vs. L.A. Abrasion

3.10

POOR
3.00 Specific Gravity (Apparent) 2.90

FAIR GOOD

2.80

2.70 2.60

2.50 10 20 30 40 LA Abrasion, % Loss 50 60 70

Figure O.39: Specific Gravity (Apparent) vs. L.A. Abrasion

422

4.50

POOR
4.00 Particle Shape Factor 3.50 3.00 2.50 2.00 1.50 1.00 10 20 30 40 LA Abrasion, % Loss 50 60 70

FAIR GOOD

Figure O.40: Particle Shape Factor vs. L.A. Abrasion

45 Aggregate Crushing Value, % Loss 40 35 30 25 20 15 10 5 0 0.0 5.0 10.0 15.0 20.0 25.0 Canadian Freeze-Thaw , % Loss

POOR FAIR GOOD

Figure 0.41: Aggregate Crushing Value vs. Canadian Freeze-Thaw Soundness

423

45 Aggregate Crushing Value (SSD), % Loss 40 35 30 25 20 15 10 0.0 5.0 10.0 15.0 20.0 25.0 Canadian Freeze-Thaw , % Loss

POOR FAIR GOOD

Figure O.42: Aggregate Crushing Value (SSD) vs. Canadian Freeze-Thaw Soundness

6.0 5.0 4.0 Absorption 3.0 2.0

POOR
1.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 Canadian Freeze-Thaw , % Loss

FAIR GOOD

Figure O.43: Absorption vs. Canadian Freeze-Thaw Soundness

424

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 Canadian Freeze-Thaw , % Loss

POOR FAIR GOOD

Figure O.44: Specific Gravity (Bulk) vs. Canadian Freeze-Thaw Soundness

3.10 3.00 Specific Gravity (SSD) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 0.0 5.0 10.0 15.0 20.0 25.0 Canadian Freeze-Thaw , % Loss

POOR FAIR GOOD

Figure O.45: Specific Gravity (SSD) vs. Canadian Freeze-Thaw Soundness

425

3.10

3.00 Specific Gravity (Apparent) 2.90

2.80 2.70

POOR FAIR GOOD

2.60 2.50 0.0 5.0 10.0 Freeze-Thaw15.0 Canadian , % Loss 20.0

25.0

Figure O.46: Specific Gravity (Apparent) vs. Canadian Freeze-Thaw Soundness

4.50

POOR
4.00 Particle Shape Factor 3.50 3.00 2.50 2.00 1.50 1.00 0.0 5.0 10.0 15.0 20.0 25.0 Canadian Freeze-Thaw , % Loss

FAIR GOOD

Figure O.47: Particle Shape Factor vs. Canadian Freeze-Thaw Soundness

426

45 Aggregate Crushing Value (SSD), % Loss 40 35 30 25 20 15 10 10 15 20 25 30 35 40 45 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

Figure O.48: Aggregate Crushing Value (SSD) vs. Aggregate Crushing Value

6.0 5.0 4.0 Absorption 3.0 2.0

POOR
1.0 0.0 10 15 20 25 30 35 40 45 Aggregate Crushing Value, % Loss

FAIR GOOD

Figure O.49: Absorption vs. Aggregate Crushing Value

427

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 10 15 20 25 30 35 40 45 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

Figure O.50: Specific Gravity (Bulk) vs. Aggregate Crushing Value

3.10

POOR
3.00 Specific Gravity (SSD) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 10 15 20 25 30 35 40 45 Aggregate Crushing Value, % Loss

FAIR GOOD

Figure O.51: Specific Gravity (SSD) vs. Aggregate Crushing Value

428

3.10

POOR
3.00 Specific Gravity (Apparent)

FAIR GOOD

2.90 2.80 2.70

2.60 2.50 10 15 20 25 30 35 40 45 Aggregate Crushing Value, % Loss

Figure O.52: Specific Gravity (Apparent) vs. Aggregate Crushing Value

4.50

POOR
4.00 Particle Shape Factor 3.50 3.00 2.50 2.00 1.50 1.00 10 15 20 25 30 35 40 45 Aggregate Crushing Value, % Loss

FAIR GOOD

Figure O.53: Particle Shape Factor vs. Aggregate Crushing Value

429

6.0 5.0 4.0 Absorption 3.0 2.0

POOR
1.0 0.0 10 15 20 25 30 35 40 45 Aggregate Crushing Value (SSD), % Loss

FAIR GOOD

Figure O.54: Absorption vs. Aggregate Crushing Value (SSD)

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 10 15 20 25 30 35 40 45 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

Figure O.55: Specific Gravity (Bulk) vs. Aggregate Crushing Value (SSD)

430

3.10

POOR
3.00 Specific Gravity (SSD) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 10 15 20 25 30 35 40 45 Aggregate Crushing Value (SSD), % Loss

FAIR GOOD

Figure O.56: Specific Gravity (SSD) vs. Aggregate Crushing Value (SSD)

3.10

POOR
3.00 Specific Gravity (Apparent) 2.90

FAIR GOOD

2.80 2.70 2.60 2.50 10 15 20 25 30 35 40 45 Aggregate Crushing Value (SSD), % Loss

Figure O.57: Specific Gravity (Apparent) vs. Aggregate Crushing Value (SSD)

431

4.50

POOR
4.00 Particle Shape Factor 3.50 3.00 2.50 2.00 1.50 1.00 10 15 20 25 30 35 40 45 Aggregate Crushing Value (SSD), % Loss

FAIR GOOD

Figure O.58: Particle Shape Factor vs. Aggregate Crushing Value (SSD)

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 1.0 2.0 3.0 Absorption 4.0 5.0 6.0

POOR FAIR GOOD

Figure O.59: Specific Gravity (Bulk) vs. Absorption

432

3.10

POOR
3.00 Specific Gravity (SSD) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 0.0 1.0 2.0 3.0 Absorption 4.0 5.0 6.0

FAIR GOOD

Figure O.60: Specific Gravity (SSD) vs. Absorption

3.10

POOR
3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 0.0 1.0 2.0 3.0 Absorption 4.0 5.0 6.0

FAIR GOOD

Figure O.61: Specific Gravity (Apparent) vs. Absorption

433

4.50

POOR
4.00 Particle Shape Factor 3.50 3.00 2.50 2.00 1.50 1.00 0.0 1.0 2.0 3.0 Absorption 4.0 5.0 6.0

FAIR GOOD

Figure O.62: Particle Shape Factor vs. Absorption

3.10 3.00 Specific Gravity (SSD) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10

Specific Gravity (Bulk)

Figure O.63: Specific Gravity (SSD) vs. Specific Gravity (Bulk)

434

3.10 3.00 Specific Gravity (Apparent) 2.90 2.80 2.70

POOR
2.60 2.50 2.20

FAIR GOOD
2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10

Specific Gravity (Bulk)

Figure O.64: Specific Gravity (Apparent) vs. Specific Gravity (Bulk)

4.50

POOR
4.00 Particle Shape Factor 3.50 3.00 2.50 2.00 1.50 1.00 2.20

FAIR GOOD

2.30

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (Bulk)

Figure O.65: Particle Shape Factor vs. Specific Gravity (Bulk)

435

3.10 3.00 Specific Gravity (Apparent) 2.90 2.80 2.70

POOR
2.60 2.50 2.30

FAIR GOOD
2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10

Specific Gravity (SSD)

Figure O.66: Specific Gravity (Apparent) vs. Specific Gravity (SSD)

4.50

POOR
4.00 Particle Shape Factor 3.50 3.00 2.50 2.00 1.50 1.00 2.30

FAIR GOOD

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (SSD)

Figure O.67: Particle Shape Factor vs. Specific Gravity (SSD)

436

4.50

POOR
4.00 Particle Shape Factor 3.50 3.00 2.50 2.00 1.50 1.00 2.50

FAIR GOOD

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (Apparent)

Figure O.68: Particle Shape Factor vs. Specific Gravity (Apparent)

437

Appendix P:

Open-Graded Friction Course Aggregates Graphs

439

30.0

25.0

Micro-Deval, % Loss

20.0

15.0

10.0

5.0

0.0

Poor

Fair
Performance

Good

Figure P.1: Micro-Deval vs. Performance

80.0 Magnesium Sulfate Soundness, % Loss 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0

Poor

Fair
Performance

Good

Figure P.2: Magnesium Sulfate Soundness vs. Performance

440

70 60 50 40

LA Abrasion, % Loss

30 20 10 0

Poor

Fair Performance

Good

Figure P.3: L.A. Abrasion vs. Performance

10.0 9.0 Canadian Freeze-Thaw, % Loss 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0

Poor

Fair
Performance

Good

Figure P.4: Canadian Freeze-Thaw vs. Performance

441

35 30 Aggregate Crushing Value, % Loss 25 20 15 10 5 0

Poor

Fair
Performance

Good

Figure P.5: Aggregate Crushing Value vs. Performance

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0

Poor

Fair
Performance

Good

Figure P.6: Aggregate Crushing Value (SSD) vs. Performance

442

3.0

2.5

2.0 Absorption, %

1.5

1.0

0.5

0.0

Poor

Fair
Performance

Good

Figure P.7: Absorption vs. Performance

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20

Poor

Fair
Performance

Good

Figure P.8: Specific Gravity (Bulk) vs. Performance

443

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 Performance

Figure P.9: Specific Gravity (SSD) vs. Performance

3.10 3.00 2.90 Specific Gravity (Apparent) 2.80 2.70 2.60 2.50 2.40 2.30

Poor
2.20

Fair

Good

Performance

Figure P.10:Specific Gravity (Apparent) vs. Performance

444

3.50 3.00 2.50 Particle Shape Factor 2.00

1.50 1.00 0.50 0.00

Poor

Fair
Performance

Good

Figure P.11:Particle Shape Factor vs. Performance

POOR % of total 100 90 - 99 80 - 89 70 - 79 60 - 69 50 - 59 40 - 49 30 - 39 20 - 29 10 - 19 0 - 10 number 2 1 0 0 0 0 0 0 0 0 0 overall % 67% 33% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Fractured Particles (1 or more sides) FAIR GOOD number 2 0 0 0 0 0 0 0 0 0 0 overall % 100% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% number 13 3 1 0 0 0 0 0 0 0 0 overall % 76% 18% 6% 0% 0% 0% 0% 0% 0% 0% 0% POOR 12% 25% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Overall % FAIR 12% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% GOOD 76% 75% 100% 0% 0% 0% 0% 0% 0% 0% 0%

Table N.1: Fractured Particles (1 or more sides) vs. Performance

445

POOR % of total 100 90 - 99 80 - 89 70 - 79 60 - 69 50 - 59 40 - 49 30 - 39 20 - 29 10 - 19 0 - 10 number 2 1 0 0 0 0 0 0 0 0 0 overall % 67% 33% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Fractured Particles (2 or more sides) FAIR GOOD number 2 0 0 0 0 0 0 0 0 0 0 overall % 100% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% number 13 3 0 1 0 0 0 0 0 0 0 overall % 76% 18% 0% 6% 0% 0% 0% 0% 0% 0% 0% POOR 12% 25% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Overall % FAIR 12% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% GOOD 76% 75% 0% 100% 0% 0% 0% 0% 0% 0% 0%

Table N.2: Fractured Particles (2 or more sides) vs. Performance

80.0 Magnesium Sulfate Soundness, % Loss 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

POOR FAIR GOOD

Figure P.12:Magnesium Sulfate Soundness vs. Micro-Deval

446

70 60 50 40 30 20 10 0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

POOR FAIR GOOD

LA Abrasion, % Loss

Figure P.13:L.A. Abrasion vs. Micro-Deval

10.0 9.0 Canadian Freeze-Thaw, % Loss 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

POOR FAIR GOOD

30.0

Figure P.14:Canadian Freeze-Thaw vs. Micro-Deval

447

35 30 25 20 15 10 5 0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

Aggregate Crushing Value, % Loss

POOR FAIR GOOD

30.0

Figure P.15:Aggregate Crushing Value vs. Micro-Deval

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

POOR FAIR GOOD

Figure P.16:Aggregate Crushing Value (SSD) vs. Micro-Deval 448

2.5

2.0

Absorption, %

1.5

1.0

0.5

0.0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

POOR FAIR GOOD

30.0

Figure P.17:Absorption vs. Micro-Deval

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

POOR FAIR GOOD

30.0

Figure P.18:Specific Gravity (Bulk) vs. Micro-Deval 449

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

POOR FAIR GOOD

30.0

Figure P.19:Specific Gravity (SSD) vs. Micro-Deval

3.10 3.00 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

Specific Gravity (Apparent)

POOR FAIR GOOD

30.0

Figure P.20:Specific Gravity (Apparent) vs. Micro-Deval 450

3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

Particle Shape Factor

POOR FAIR GOOD

30.0

Figure P.21:Particle Shape Factor vs. Micro-Deval

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0 30.0

POOR FAIR GOOD

Figure P.22:Fractured Particles (1 or more sides) vs. Micro-Deval 451

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 5.0 10.0 15.0 Micro-Deval, % Loss 20.0 25.0

POOR FAIR GOOD

30.0

Figure P.23:Fractured Particles (2 or more sides) vs. Micro-Deval

70 60 50 40 30 20 10 0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

LA Abrasion, % Loss

POOR FAIR GOOD

80.0

Figure P.24:L.A. Abrasion vs. Magnesium Sulfate Soundness

452

10.0 9.0 Canadian Freeze-Thaw, % Loss 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

80.0

Figure P.25:Canadian Freeze-Thaw vs. Magnesium Sulfate Soundness

35 30 25 20 15 10 5 0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

Aggregate Crushing Value, % Loss

POOR FAIR GOOD

80.0

Figure P.26:Aggregate Crushing Value vs. Magnesium Sulfate Soundness

453

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

80.0

Figure P.27:Aggregate Crushing Value (SSD) vs. Magnesium Sulfate Soundness

3.0

2.5

2.0 Absorption, %

1.5

1.0

0.5

0.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

80.0

Figure P.28:Absorption vs. Magnesium Sulfate Soundness

454

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

80.0

Figure P.29:Specific Gravity (Bulk) vs. Magnesium Sulfate Soundness

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

80.0

Figure P.30:Specific Gravity (SSD) vs. Magnesium Sulfate Soundness 455

3.10 3.00 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

Specific Gravity (Apparent)

POOR FAIR GOOD

80.0

Figure P.31:Specific Gravity (Apparent) vs. Magnesium Sulfate Soundness

3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

Particle Shape Factor

POOR FAIR GOOD

80.0

Figure P.32:Particle Shape Factor vs. Magnesium Sulfate Soundness

456

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

80.0

Figure P.33:Fractured Particles (1 or more sides) vs. Magnesium Sulfate Soundness

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Magnesium Sulfate Soundness, % Loss

POOR FAIR GOOD

80.0

Figure P.34:Fractured Particles (2 or more sides) vs. Magnesium Sulfate Soundness

457

10.0 9.0 Canadian Freeze-Thaw, % Loss 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 0 10 20 30 40 50 60 LA Abrasion, % Loss

POOR FAIR GOOD

70

Figure P.35:Canadian Freeze-Thaw vs. L.A. Abrasion

35 30 25 20 15 10 5 0 0 10 20 30 40 50 60 LA Abrasion, % Loss

Aggregate Crushing Value, % Loss

POOR FAIR GOOD

70

Figure P.36:Aggregate Crushing Value vs. L.A. Abrasion 458

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0 10 20 30 40 50 60 LA Abrasion, % Loss

POOR FAIR GOOD

70

Figure P.37:Aggregate Crushing Value (SSD) vs. L.A. Abrasion

3.0

2.5

2.0 Absorption, %

1.5

1.0

0.5

0.0 0 10 20 30 40 50 60 LA Abrasion, % Loss

POOR FAIR GOOD

70

Figure P.38:Absorption vs. L.A. Abrasion 459

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0 10 20 30 40 50 60 LA Abrasion, % Loss

POOR FAIR GOOD

70

Figure P.39:Specific Gravity (Bulk) vs. L.A. Abrasion

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0 10 20 30 40 50 60 LA Abrasion, % Loss

POOR FAIR GOOD

70

Figure P.40:Specific Gravity (SSD) vs. L.A. Abrasion

460

3.10 3.00 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0 10 20 30 40 50 60 70 LA Abrasion, % Loss

Specific Gravity (Apparent)

POOR FAIR GOOD

Figure P.41:Specific Gravity (Apparent) vs. L.A. Abrasion

3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0 10 20 30 40 50 60 LA Abrasion, % Loss

Particle Shape Factor

POOR FAIR GOOD

70

Figure P.42:Particle Shape Factor vs. L.A. Abrasion

461

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 LA Abrasion, % Loss

POOR FAIR GOOD

70

Figure P.43:Fractured Particles (1 or more sides) vs. L.A. Abrasion

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 LA Abrasion, % Loss

POOR FAIR GOOD

70

Figure P.44:Fractured Particles (2 or more sides) vs. L.A. Abrasion

462

35 30 25 20 15 10 5 0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Canadian Freeze-Thaw, % Loss

Aggregate Crushing Value, % Loss

POOR FAIR GOOD

9.0 10.0

Figure P.45:Aggregate Crushing Value vs. Canadian Freeze-Thaw

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

9.0 10.0

Figure P.46:Aggregate Crushing Value (SSD) vs. Canadian Freeze-Thaw

463

3.0

2.5

2.0 Absorption, %

1.5

1.0

0.5

0.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

9.0 10.0

Figure P.47:Absorption vs. Canadian Freeze-Thaw

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

9.0 10.0

Figure P.48:Specific Gravity (Bulk) vs. Canadian Freeze-Thaw

464

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

9.0 10.0

Figure P.49:Specific Gravity (SSD) vs. Canadian Freeze-Thaw

3.10 3.00 2.90 Specific Gravity (Apparent) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

9.0 10.0

Figure P.50:Specific Gravity (Apparent) vs. Canadian Freeze-Thaw

465

3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Canadian Freeze-Thaw, % Loss

Particle Shape Factor

POOR FAIR GOOD

9.0 10.0

Figure P.51:Particle Shape Factor vs. Canadian Freeze-Thaw

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

9.0 10.0

Figure P.52:Fractured Particles (1 or more sides) vs. Canadian Freeze-Thaw

466

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Canadian Freeze-Thaw, % Loss

POOR FAIR GOOD

9.0 10.0

Figure P.53:Fractured Particles (2 or more sides) vs. Canadian Freeze-Thaw

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0 5 10 15 20 25 30 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

35

Figure P.54:Aggregate Crushing Value (SSD) vs. Aggregate Crushing Value

467

3.0

2.5

2.0 Absorption, %

1.5

1.0

0.5

0.0 0 5 10 15 20 25 30 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

35

Figure P.55:Absorption vs. Aggregate Crushing Value

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0 5 10 15 20 25 30 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

35

Figure P.56:Specific Gravity (Bulk) vs. Aggregate Crushing Value

468

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0 5 10 15 20 25 30 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

35

Figure P.57:Specific Gravity (SSD) vs. Aggregate Crushing Value

3.10 3.00 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0 5 10 15 20 25 30 Aggregate Crushing Value, % Loss

Specific Gravity (Apparent)

POOR FAIR GOOD

35

Figure P.58:Specific Gravity (Apparent) vs. Aggregate Crushing Value

469

3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0 5 10 15 20 25 30 Aggregate Crushing Value, % Loss

Particle Shape Factor

POOR FAIR GOOD

35

Figure P.59:Particle Shape Factor vs. Aggregate Crushing Value

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 5 10 15 20 25 30 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

35

Figure P.60:Fractured Particles (1 or more sides) vs. Aggregate Crushing Value

470

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 5 10 15 20 25 30 Aggregate Crushing Value, % Loss

POOR FAIR GOOD

35

Figure P.61:Fractured Particles (2 or more sides) vs. Aggregate Crushing Value

3.0

2.5

2.0 Absorption, %

1.5

1.0

0.5

0.0 0 5 10 15 20 25 30 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

35

Figure P.62:Absorption vs. Aggregate Crushing Value (SSD)

471

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0 5 10 15 20 25 30 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

35

Figure P.63:Specific Gravity (Bulk) vs. Aggregate Crushing Value (SSD)

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0 5 10 15 20 25 30 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

35

Figure P.64:Specific Gravity (SSD) vs. Aggregate Crushing Value (SSD)

472

3.10 3.00 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0 5 10 15 20 25 30 Aggregate Crushing Value (SSD), % Loss

Specific Gravity (Apparent)

POOR FAIR GOOD

35

Figure P.65:Specific Gravity (Apparent) vs. Aggregate Crushing Value (SSD)

3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0 5 10 15 20 25 30 Aggregate Crushing Value (SSD), % Loss

Particle Shape Factor

POOR FAIR GOOD

35

Figure P.66:Particle Shape Factor vs. Aggregate Crushing Value (SSD)

473

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 5 10 15 20 25 30 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

35

Figure P.67:Fractured Particles (1 or more sides) vs. Aggregate Crushing Value (SSD)
100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 5 10 15 20 25 30 Aggregate Crushing Value (SSD), % Loss

POOR FAIR GOOD

35

Figure P.68:Fractured Particles (2 or more sides) vs. Aggregate Crushing Value (SSD)

474

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 0.5 1.0 1.5 Absorption, % 2.0 2.5

POOR FAIR GOOD

3.0

Figure P.69:Specific Gravity (Bulk) vs. Absorption

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 0.5 1.0 1.5 Absorption, % 2.0 2.5

POOR FAIR GOOD

3.0

Figure P.70:Specific Gravity (SSD) vs. Absorption 475

3.10 3.00 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 0.5 1.0 1.5 Absorption, % 2.0 2.5

Specific Gravity (Apparent)

POOR FAIR GOOD

3.0

Figure P.71:Specific Gravity (Apparent) vs. Absorption

3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 0.0 0.5 1.0 1.5 Absorption, % 2.0 2.5

Particle Shape Factor

POOR FAIR GOOD

3.0

Figure P.72:Particle Shape Factor vs. Absorption 476

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 0.5 1.0 1.5 Absorption, % 2.0 2.5

POOR FAIR GOOD

3.0

Figure P.73:Fractured Particles (1 or more sides) vs. Absorption

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0.0 0.5 1.0 1.5 Absorption, % 2.0 2.5

POOR FAIR GOOD

3.0

Figure P.74:Fractured Particles (2 or more sides) vs. Absorption 477

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10 Specific Gravity (Bulk)

Figure P.75:Specific Gravity (SSD) vs. Specific Gravity (Bulk)

3.10 3.00 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20 2.20

Specific Gravity (Apparent)

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10

Specific Gravity (Bulk)

Figure P.76:Specific Gravity (Apparent) vs. Specific Gravity (Bulk)

478

3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 2.20

Particle Shape Factor

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10 Specific Gravity (Bulk)

Figure P.77:Particle Shape Factor vs. Specific Gravity (Bulk)

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10 Specific Gravity (Bulk)

Figure P.78:Fractured Particles (1 or more sides) vs. Specific Gravity (Bulk)

479

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10 Specific Gravity (Bulk)

Figure P.79:Fractured Particles (2 or more sides) vs. Specific Gravity (Bulk)

3.10 3.00 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20 2.20

Specific Gravity (Apparent)

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10 Specific Gravity (SSD)

Figure P.80:Specific Gravity (Apparent) vs. Specific Gravity (SSD)

480

3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 2.20

Particle Shape Factor

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10 Specific Gravity (SSD)

Figure P.81:Particle Shape Factor vs. Specific Gravity (SSD)

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10 Specific Gravity (SSD)

Figure P.82:Fractured Particles (1 or more sides) vs. Specific Gravity (SSD)

481

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10 Specific Gravity (SSD)

Figure P.83:Fractured Particles (2 or more sides) vs. Specific Gravity (SSD)

3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 2.20

Particle Shape Factor

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10 Specific Gravity (Apparent)

Figure P.84:Particle Shape Factor vs. Specific Gravity (Apparent)

482

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10 Specific Gravity (Apparent)

Figure P.85:Fractured Particles (1 or more sides) vs. Specific Gravity (Apparent)

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 2.20

POOR FAIR GOOD

2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 3.10 Specific Gravity (Apparent)

Figure P.86:Fractured Particles (2 or more sides) vs. Specific Gravity (Apparent)

483

100 Fractured Particles (1 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 1.00

POOR FAIR GOOD

1.50 2.00 2.50 3.00 3.50 Particle Shape Factor

Figure P.87:Fractured Particles (1 or more sides) vs. Particle Shape Factor

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 1.00

POOR FAIR GOOD

1.50 2.00 2.50 3.00 3.50 Particle Shape Factor

Figure P.88:Fractured Particles (2 or more sides) vs. Particle Shape Factor

484

100 Fractured Particles (2 or more sides), % of total 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 Fractured Particles (1 or more sides), % of total

POOR FAIR GOOD

90 100

Figure P.89:Fractured Particles (2 or more sides) vs. Fractured Particles (1 or more sides)

485

Appendix Q:

Test Correlation Graphs for the Full Data Set

486

80.0 Magnesium Sulfate Soundness, % Loss 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 0.0 10.0 20.0 30.0 Micro-Deval, % Loss 40.0 50.0 60.0

y = 0.0241x 2 + 0.1969x + 0.5049 R2 = 0.5362

Figure Q.1:

Magnesium Sulfate Soundness vs. Micro-Deval

70 60 LA Abrasion, % Loss 50 40 30 20 10 0 0.0 10.0 20.0 30.0 Micro-Deval, % Loss 40.0 50.0 60.0

y = 14.779x 0.2101 R2 = 0.116

Figure Q.2:

L.A. Abrasion vs. Micro-Deval

487

25.0

Canadian Freeze-Thaw, % Loss

20.0

y = 0.0042x 2 + 0.07x + 2.2535 R2 = 0.3204

15.0

10.0

5.0

0.0 0.0 10.0 20.0 30.0 Micro-Deval, % Loss 40.0 50.0 60.0

Figure Q.3:

Canadian Freeze-Thaw vs. Micro-Deval

60 Aggregate Crushing Value, % Loss 50 40

y = 12.07x 0.2093 R2 = 0.2196

30 20 10 0 0.0 10.0 20.0 30.0 Micro-Deval, % Loss 40.0 50.0 60.0

Figure Q.4:

Aggregate Crushing Value vs. Micro-Deval

488

60.00 Aggregate Crushing Value (SSD), % Loss 50.00

y = 12.537x 0.2046 R2 = 0.196

40.00 30.00 20.00

10.00 0.00 0.0 10.0 20.0 30.0 Micro-Deval, % Loss 40.0 50.0 60.0

Figure Q.5:

Aggregate Crushing Value (SSD) vs. Micro-Deval

7.0 6.0 5.0 Absorption 4.0 3.0 2.0 1.0 0.0 0.0

y = 0.0019x 2 + 0.017x + 0.6582 R2 = 0.4005

10.0

20.0

30.0 Micro-Deval, % Loss

40.0

50.0

60.0

Figure Q.6:

Absorption vs. Micro-Deval

489

3.50 3.00 Specific Gravity (Bulk) 2.50 2.00 1.50 1.00 0.50 0.00 0.0 10.0 20.0

y = -0.0002x 2 + 0.0007x + 2.6571 R2 = 0.1718

30.0 Micro-Deval, % Loss

40.0

50.0

60.0

Figure Q.7:

Specific Gravity (Bulk) vs. Micro-Deval

3.50 3.00 Specific Gravity (SSD) 2.50 2.00 1.50 1.00 0.50 0.00 0.0 10.0 20.0 30.0 Micro-Deval, % Loss 40.0 50.0 60.0

y = -0.0001x 2 + 0.0011x + 2.6751 R2 = 0.1135

Figure Q.8:

Specific Gravity (SSD) vs. Micro-Deval

490

3.50 3.00 Specific Gravity (Apparent) 2.50 2.00 1.50 1.00 0.50 0.00 0.0 10.0 20.0 30.0 Micro-Deval, % Loss 40.0 50.0 60.0

y = -7E-05x 2 + 0.0019x + 2.7049 R2 = 0.0124

Figure Q.9:

Specific Gravity (Apparent) vs. Micro-Deval

6.00

5.00 Particle Shape Factor

y = 0.1873Ln(x) + 1.8586 R2 = 0.0371

4.00

3.00

2.00

1.00

0.00 0.0 10.0 20.0 30.0 Micro-Deval, % Loss 40.0 50.0 60.0

Figure Q.10: Particle Shape Factor vs. Micro-Deval

491

70 60 LA Abrasion, % Loss 50 40 30 20 10 0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

y = 0.0018x 2 + 0.2568x + 24.036 R2 = 0.1785

Figure Q.11: L.A. Abrasion vs. Magnesium Sulfate Soundness

25.0

Canadian Freeze-Thaw, % Loss

20.0

15.0

y = 1.8391x 0.3782 R2 = 0.3887

10.0

5.0

0.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

Figure Q.12: Canadian Freeze-Thaw vs. Magnesium Sulfate Soundness

492

60 Aggregate Crushing Value, % Loss 50 40 30 20 10 0 0.0 10.0 20.0 30.0 40.0 50.0

y = 17.662x 0.0827 R2 = 0.1241

60.0

70.0

80.0

Magnesium Sulfate Soundness, % Loss

Figure Q.13: Aggregate Crushing Value vs. Magnesium Sulfate Soundness

60.00 Aggregate Crushing Value (SSD), % Loss 50.00 40.00

y = 1.4869Ln(x) + 19.141 R2 = 0.0799

30.00 20.00 10.00 0.00 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

Figure Q.14: Aggregate Crushing Value (SSD) vs. Magnesium Sulfate Soundness

493

6.0 5.0

y = 0.4224x 0.5258 R2 = 0.5358

4.0 Absorption 3.0 2.0

1.0 0.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

Figure Q.15: Absorption vs. Magnesium Sulfate Soundness

3.50 3.00 Specific Gravity (Bulk) 2.50 2.00 1.50 1.00 0.50 0.00 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

y = -0.0669Ln(x) + 2.7301 R2 = 0.3111

Figure Q.16: Specific Gravity (Bulk) vs. Magnesium Sulfate Soundness

494

3.50 3.00 Specific Gravity (SSD) 2.50 2.00 1.50 1.00 0.50 0.00 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

y = -0.0504Ln(x) + 2.738 R2 = 0.2471

Figure Q.17: Specific Gravity (SSD) vs. Magnesium Sulfate Soundness

3.50 3.00 Specific Gravity (Apparent) 2.50 2.00 1.50 1.00 0.50 0.00 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

y = 2.752x -0.0086 R2 = 0.0786

Figure Q.18: Specific Gravity (Apparent) vs. Magnesium Sulfate Soundness

495

6.00

5.00 Particle Shape Factor

4.00

y = -0.0002x 2 + 0.0129x + 2.2639 R2 = 0.0131

3.00

2.00

1.00

0.00 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Magnesium Sulfate Soundness, % Loss

Figure Q.19: Particle Shape Factor vs. Magnesium Sulfate Soundness

25.0

Canadian Freeze-Thaw, % Loss

20.0

y = -0.0036x 2 + 0.2265x + 1.5103 R2 = 0.0409

15.0

10.0

5.0

0.0 0 10 20 30 40 50 60 70 LA Abrasion, % Loss

Figure Q.20: Canadian Freeze-Thaw vs. L.A. Abrasion

496

50 Aggregate Crushing Value, % Loss 45 40 35 30 25 20 15 10 5 0 10 20 30

y = 3.1264x 0.5835 R2 = 0.6495

40 LA Abrasion, % Loss

50

60

70

Figure Q.21: Aggregate Crushing Value vs. L.A. Abrasion

60.00 Aggregate Crushing Value (SSD), % Loss 50.00 40.00 30.00 20.00 10.00 0.00 0 10 20 30 40 50 60 70 LA Abrasion, % Loss

y = -0.0054x 2 + 0.7696x + 5.7461 R2 = 0.4868

Figure Q.22: Aggregate Crushing Value (SSD) vs. L.A. Abrasion

497

6.0 5.0 4.0 Absorption

y = -0.0011x 2 + 0.0991x - 0.2811 R2 = 0.0749

3.0 2.0 1.0 0.0 10 20 30 40 LA Abrasion, % Loss 50 60 70

Figure Q.23: Absorption vs. L.A. Abrasion

3.20 3.00 Specific Gravity (Bulk)

y = 0.0002x 2 - 0.0154x + 2.8926 R2 = 0.1707

2.80 2.60 2.40

2.20 2.00 10 20 30 40 LA Abrasion, % Loss 50 60 70

Figure Q.24: Specific Gravity (Bulk) vs. L.A. Abrasion

498

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 10 20 30 40 LA Abrasion, % Loss 50 60 70

y = 0.0001x 2 - 0.0128x + 2.8856 R2 = 0.1798

Figure Q.25: Specific Gravity (SSD) vs. L.A. Abrasion

3.10 3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 10 20 30 40 LA Abrasion, % Loss 50 60 70

y = 9E-05x 2 - 0.0093x + 2.8889 R2 = 0.1683

Figure Q.26: Specific Gravity (Apparent) vs. L.A. Abrasion

499

6.00

5.00 Particle Shape Factor

y = 2.2877e-0.0003x R2 = 0.0002

4.00

3.00 2.00

1.00

0.00 10 20 30 40 LA Abrasion, % Loss 50 60 70

Figure Q.27: Particle Shape Factor vs. L.A. Abrasion

50 Aggregate Crushing Value, % Loss 45 40 35 30 25 20 15 10 0.0 5.0 10.0 15.0 20.0 25.0 Canadian Freeze-Thaw , % Loss

y = 19.75e0.0087x R2 = 0.0119

Figure Q.28: Aggregate Crushing Value vs. Canadian Freeze-Thaw

500

60.00 Aggregate Crushing Value (SSD), % Loss 50.00 40.00 30.00 20.00 10.00 0.00 0.0 5.0 10.0 15.0

y = 20.503e0.006x R2 = 0.0053

20.0

25.0

Canadian Freeze-Thaw , % Loss

Figure Q.29: Aggregate Crushing Value (SSD) vs. Canadian Freeze-Thaw

6.0 5.0 4.0 Absorption 3.0 2.0 1.0 0.0 0.0 5.0 10.0

y = 0.5943x 0.4538 R2 = 0.1468

15.0

20.0

25.0

Canadian Freeze-Thaw , % Loss

Figure Q.30: Absorption vs. Canadian Freeze-Thaw

501

3.20 3.00 Specific Gravity (Bulk) 2.80

y = 0.0002x 2 - 0.0098x + 2.652 R2 = 0.0312

2.60 2.40 2.20 2.00 0.0 5.0 10.0 15.0 20.0 25.0 Canadian Freeze-Thaw , % Loss

Figure Q.31: Specific Gravity (Bulk) vs. Canadian Freeze-Thaw

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 5.0 10.0 15.0 20.0 25.0 Canadian Freeze-Thaw , % Loss

y = -0.0238Ln(x) + 2.6803 R2 = 0.0202

Figure Q.32: Specific Gravity (SSD) vs. Canadian Freeze-Thaw

502

3.10 3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 0.0 5.0 10.0 15.0 20.0 25.0 Canadian Freeze-Thaw , % Loss

y = 0.0003x 2 - 0.0036x + 2.7184 R2 = 0.0089

Figure Q.33: Specific Gravity (Apparent) vs. Canadian Freeze-Thaw

6.00

5.00 Particle Shape Factor 4.00

y = 0.0013x 2 - 0.0132x + 2.3531 R2 = 0.004

3.00

2.00 1.00

0.00 0.0 5.0 10.0 15.0 20.0 25.0 Canadian Freeze-Thaw , % Loss

Figure Q.34: Particle Shape Factor vs. Canadian Freeze-Thaw

503

60.00 Aggregate Crushing Value (SSD), % Loss 50.00 40.00 30.00 20.00 10.00 0.00 10

y = 0.9539x + 1.5506 R2 = 0.836

15

20

25

30

35

40

45

50

Aggregate Crushing Value, % Loss

Figure Q.35: Aggregate Crushing Value (SSD) vs. Aggregate Crushing Value

6.0 5.0 4.0 Absorption 3.0 2.0 1.0 0.0 10 15 20 25 30

y = 0.0035x 2 - 0.1049x + 2.0038 R2 = 0.1414

35 40 45 50

Aggregate Crushing Value, % Loss

Figure Q.36: Absorption vs. Aggregate Crushing Value

504

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = -0.0003x 2 + 0.0047x + 2.6483 R2 = 0.1666

Figure Q.37: Specific Gravity (Bulk) vs. Aggregate Crushing Value

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = -0.0002x 2 + 0.0024x + 2.6972 R2 = 0.1605

Figure Q.38: Specific Gravity (SSD) vs. Aggregate Crushing Value

505

3.10 3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = -0.0001x 2 - 0.0006x + 2.7738 R2 = 0.1217

Figure Q.39: Specific Gravity (Apparent) vs. Aggregate Crushing Value

6.00

5.00 Particle Shape Factor 4.00

y = 0.0012x 2 - 0.0548x + 2.8934 R2 = 0.0329

3.00 2.00

1.00 0.00 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

Figure Q.40: Particle Shape Factor vs. Aggregate Crushing Value

506

6.0 5.0 4.0 Absorption 3.0 2.0 1.0 0.0 0

y = 0.004x 2 - 0.1366x + 2.4048 R2 = 0.1339

10

15

20

25

30

35

40

45

50

Aggregate Crushing Value (SSD), % Loss

Figure Q.41: Absorption vs. Aggregate Crushing Value (SSD)

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.00

y = -0.0005x 2 + 0.0165x + 2.5051 R2 = 0.172

5.00

10.00

15.00

20.00

25.00

30.00

35.00

40.00

45.00

50.00

Aggregate Crushing Value (SSD), % Loss

Figure Q.42: Specific Gravity (Bulk) vs. Aggregate Crushing Value (SSD)

507

3.10 3.00 Specific Gravity (SSD) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 0.00

y = -0.0004x 2 + 0.0139x + 2.558 R2 = 0.1693

5.00

10.00

15.00

20.00

25.00

30.00

35.00

40.00

45.00

50.00

Aggregate Crushing Value (SSD), % Loss

Figure Q.43: Specific Gravity (SSD) vs. Aggregate Crushing Value (SSD)

3.10 3.00 Specific Gravity (Apparent) 2.90

y = -0.0003x 2 + 0.0098x + 2.6498 R2 = 0.1334

2.80 2.70 2.60 2.50 0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

40.00

45.00

50.00

Aggregate Crushing Value (SSD), % Loss

Figure Q.44: Specific Gravity (Apparent) vs. Aggregate Crushing Value (SSD)

508

5.50 5.00 4.50 Particle Shape Factor 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.00

y = 0.0011x 2 - 0.0506x + 2.8793 R2 = 0.0278

5.00

10.00

15.00

20.00

25.00

30.00

35.00

40.00

45.00

50.00

Aggregate Crushing Value (SSD), % Loss

Figure Q.45: Particle Shape Factor vs. Aggregate Crushing Value (SSD)

3.10 3.00 2.90 Specific Gravity (Bulk) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 1.0 2.0 3.0 Absorption 4.0 5.0 6.0

y = 2.7496e-0.0357x R2 = 0.6473

Figure Q.46: Specific Gravity (Bulk) vs. Absorption

509

3.10 3.00 2.90 Specific Gravity (SSD) 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 1.0 2.0 3.0 Absorption 4.0

y = 2.7492e-0.0258x R2 = 0.4919

5.0

6.0

Figure Q.47: Specific Gravity (SSD) vs. Absorption

3.10 3.00 Specific Gravity (Apparent) 2.90 2.80 2.70 2.60 2.50 2.40 2.30 2.20 0.0 1.0 2.0 3.0 Absorption 4.0 5.0 6.0

y = 2.7134x -0.016 R2 = 0.1402

Figure Q.48: Specific Gravity (Apparent) vs. Absorption

510

6.00 5.00 Particle Shape Factor 4.00

y = 0.0219x 2 - 0.1012x + 2.4038 R2 = 0.0041

3.00 2.00 1.00 0.00 0.0 1.0 2.0 3.0 Absorption 4.0 5.0 6.0

Figure Q.49: Particle Shape Factor vs. Absorption

3.20 3.00 Specific Gravity (SSD) 2.80 2.60 2.40 2.20 2.00 2.20

y = 0.8356x + 0.4662 R2 = 0.9749

2.30

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (Bulk)

Figure Q.50: Specific Gravity (SSD) vs. Specific Gravity (Bulk)

511

3.20 3.00 Specific Gravity (Apparent) 2.80 2.60 2.40 2.20 2.00 2.20

y = 0.4688x 2 - 1.8333x + 4.2933 R2 = 0.7437

2.30

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (Bulk)

Figure Q.51: Specific Gravity (Apparent) vs. Specific Gravity (Bulk)

5.50 5.00 Particle Shape Factor 4.50 4.00 3.50 3.00 2.50 2.00 2.20

y = 1.5193x 2 - 7.6505x + 11.916 R2 = 0.0085

2.30

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (Bulk)

Figure Q.52: Particle Shape Factor vs. Specific Gravity (Bulk)

512

3.20 3.00 Specific Gravity (Apparent) 2.80 2.60 2.40 2.20 2.00 2.20

y = 0.3415x 2 - 1.0461x + 3.0821 R2 = 0.8538

2.30

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

3.20

Specific Gravity (SSD)

Figure Q.53: Specific Gravity (Apparent) vs. Specific Gravity (SSD)

6.00 5.00 Particle Shape Factor 4.00 3.00 2.00 1.00 0.00 2.20

y = 2.3252x 2 - 12.142x + 18.149 R2 = 0.0114

2.30

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (SSD)

Figure Q.54: Particle Shape Factor vs. Specific Gravity (SSD)

513

6.00 5.00 Particle Shape Factor 4.00 3.00 2.00 1.00 0.00 2.20

y = 1.1321x 0.6934 R2 = 0.0109

2.30

2.40

2.50

2.60

2.70

2.80

2.90

3.00

3.10

Specific Gravity (Apparent)

Figure Q.55: Particle Shape Factor vs. Specific Gravity (Apparent)

514

Appendix R:

Test Correlation Graphs for the Partial Data Set I

515

45 Magnesium Sulfate Soundness, % Loss 40 35 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 40 45 Micro-Deval, % Loss

y = 0.0199x 2 + 0.1805x + 0.9195 R2 = 0.5651

Figure R.1:

Magnesium Sulfate Soundness vs. Micro-Deval

70 60 LA Abrasion, % Loss 50 40 30 20 10 0 0 5 10 15 20 25 30 35 40 45 Micro-Deval, % Loss

y = 14.588x 0.2084 R2 = 0.1224

Figure R.2:

L.A. Abrasion vs. Micro-Deval

516

16 14 Canadian Freeze-Thaw, % Loss 12 10 8 6 4 2 0 0 5 10 15 20 25 30 35 40 45 Micro-Deval, % Loss

y = 1.8942e0.0418x R2 = 0.223

Figure R.3:

Canadian Freeze-Thaw vs. Micro-Deval

35 Aggregate Crushing Value, % Loss 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 40 45 Micro-Deval, % Loss

y = 13.055x 0.1714 R2 = 0.1712

Figure R.4:

Aggregate Crushing Value vs. Micro-Deval

517

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 40 45 Micro-Deval, % Loss

y = 13.451x 0.1701 R2 = 0.1543

Figure R.5:

Aggregate Crushing Value (SSD) vs. Micro-Deval

5 4 Absorption

y = 0.0046x 2 - 0.0838x + 1.4024 R2 = 0.33

2 1

0 0 5 10 15 20 25 30 35 40 45 Micro-Deval, % Loss

Figure R.6:

Absorption vs. Micro-Deval

518

3.5 3 Specific Gravity (Bulk) 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 45 Micro-Deval, % Loss

y = -0.0005x 2 + 0.0121x + 2.5842 R2 = 0.2107

Figure R.7:

Specific Gravity (Bulk) vs. Micro-Deval

3.5 3 Specific Gravity (SSD) 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 45 Micro-Deval, % Loss

y = -0.0004x 2 + 0.0112x + 2.6085 R2 = 0.1898

Figure R.8:

Specific Gravity (SSD) vs. Micro-Deval

519

3 2.95 2.9 Specific Gravity (Apparent) 2.85 2.8 2.75 2.7 2.65 2.6 2.55 2.5 2.45 0 5 10 15 20

y = -0.0003x 2 + 0.0094x + 2.6622 R2 = 0.1316

25

30

35

40

45

Micro-Deval, % Loss

Figure R.9:

Specific Gravity (Apparent) vs. Micro-Deval

4.5 4 3.5 Particle Shape Factor 3 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 45 Micro-Deval, % Loss

y = 0.1416Ln(x) + 1.9279 R2 = 0.0289

Figure R.10: Particle Shape Factor vs. Micro-Deval

520

70 60 LA Abrasion, % Loss 50 40 30 20 10 0 0 5 10 15 20 25 30 35 40 45 Magnesium Sulfate Soundness, % Loss

y = 20.067x 0.1197 R2 = 0.1293

Figure R.11: L.A. Abrasion vs. Magnesium Sulfate Soundness

16 14 Canadian Freeze-Thaw, % Loss 12 10 8 6 4 2 0 0 5 10 15 20 25 30 35 40 45 Magnesium Sulfate Soundness, % Loss

y = 1.8026x 0.3947 R2 = 0.3934

Figure R.12: Canadian Freeze-Thaw vs. Magnesium Sulfate Soundness

521

35 Aggregate Crushing Value, % Loss 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 40 45 Magnesium Sulfate Soundness, % Loss

y = 17.991x 0.0633 R2 = 0.0739

Figure R.13: Aggregate Crushing Value vs. Magnesium Sulfate Soundness

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 40 45 Magnesium Sulfate Soundness, % Loss

y = 19.061x 0.0435 R2 = 0.0321

Figure R.14: Aggregate Crushing Value (SSD) vs. Magnesium Sulfate Soundness

522

5 4 Absorption

y = 0.4134x 0.5331 R2 = 0.528

3 2

1 0 0 5 10 15 20 25 30 35 40 45 Magnesium Sulfate Soundness, % Loss

Figure R.15: Absorption vs. Magnesium Sulfate Soundness

3.5 3 Specific Gravity (Bulk) 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 45 Magnesium Sulfate Soundness, % Loss

y = -0.0665Ln(x) + 2.7296 R2 = 0.2792

Figure R.16: Specific Gravity (Bulk) vs. Magnesium Sulfate Soundness

523

3.5 3 Specific Gravity (SSD) 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 45 Magnesium Sulfate Soundness, % Loss

y = -0.045Ln(x) + 2.7293 R2 = 0.2182

Figure R.17: Specific Gravity (SSD) vs. Magnesium Sulfate Soundness

3 2.95 2.9 Specific Gravity (Apparent) 2.85 2.8 2.75 2.7 2.65 2.6 2.55 2.5 2.45 0 5 10 15 20 25 30

y = 2.7461x -0.0078 R2 = 0.0883

35

40

45

Magnesium Sulfate Soundness, % Loss

Figure R.18: Specific Gravity (Apparent) vs. Magnesium Sulfate Soundness

524

4.5 4 3.5 Particle Shape Factor 3 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 45 Magnesium Sulfate Soundness, % Loss

y = 0.0001x 2 - 0.0034x + 2.2984 R2 = 0.0012

Figure R.19: Particle Shape Factor vs. Magnesium Sulfate Soundness

16 14 Canadian Freeze-Thaw, % Loss 12 10 8 6 4 2 0 0 10 20 30 40 50 60 70 LA Abrasion, % Loss

y = -0.0039x 2 + 0.2267x + 1.5324 R2 = 0.0451

Figure R.20: Canadian Freeze-Thaw vs. L.A. Abrasion

525

35 Aggregate Crushing Value, % Loss 30 25 20 15 10 5 0 0 10 20 30 40 50 60 70 LA Abrasion, % Loss

y = 11.352Ln(x) - 15.73 R2 = 0.7044

Figure R.21: Aggregate Crushing Value vs. L.A. Abrasion

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0 10 20 30 40 50 60 70 LA Abrasion, % Loss

y = 10.736Ln(x) - 13.18 R2 = 0.5711

Figure R.22: Aggregate Crushing Value (SSD) vs. L.A. Abrasion

526

6 5

4 Absorption 3 2

y = 0.7752Ln(x) - 1.0669 R2 = 0.063

1 0 10 20 30 40 LA Abrasion, % Loss 50 60 70

Figure R.23: Absorption vs. L.A. Abrasion

3.2

3 Specific Gravity (Bulk) 2.8

2.6 2.4

2.2 2 10 20 30 40 LA Abrasion, % Loss

y = -0.1449Ln(x) + 3.0803 R2 = 0.1506

50 60 70

Figure R.24: Specific Gravity (Bulk) vs. L.A. Abrasion

527

3 2.9 Specific Gravity (SSD) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 10 20 30 40 LA Abrasion, % Loss 50 60 70

y = -0.1145Ln(x) + 3.0194 R2 = 0.1626

Figure R.25: Specific Gravity (SSD) vs. L.A. Abrasion

3 2.9 Specific Gravity (Apparent) 2.8 2.7 2.6 2.5 2.4 2.3 10 20 30 40 LA Abrasion, % Loss 50 60 70

y = -0.0782Ln(x) + 2.9635 R2 = 0.135

Figure R.26: Specific Gravity (Apparent) vs. L.A. Abrasion

528

4.5 4 3.5 Particle Shape Factor 3 2.5 2 1.5 1 0.5 0 10 20 30 40 LA Abrasion, % Loss 50 60 70

y = 0.0003x 2 - 0.017x + 2.535 R2 = 0.0061

Figure R.27: Particle Shape Factor vs. L.A. Abrasion

50 Aggregate Crushing Value, % Loss 45 40 35 30 25 20 15 10 0 2 4 6 8 10 12 14 16 Canadian Freeze-Thaw , % Loss

y = 19.322e0.0085x R2 = 0.0112

Figure R.28: Aggregate Crushing Value vs. Canadian Freeze-Thaw

529

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0 2 4 6 8 10 12 14 16 Canadian Freeze-Thaw , % Loss

y = 0.0521x 2 - 0.6043x + 22.438 R2 = 0.0145

Figure R.29: Aggregate Crushing Value (SSD) vs. Canadian Freeze-Thaw

6 5 4 Absorption

y = 0.6156x 0.4097 R2 = 0.1198

3 2 1 0 0 2 4 6 8 10 12 14 16 Canadian Freeze-Thaw , % Loss

Figure R.30: Absorption vs. Canadian Freeze-Thaw

530

3.2 3 Specific Gravity (Bulk)

y = -0.0356Ln(x) + 2.6596 R2 = 0.0308

2.8 2.6 2.4

2.2 2 0 2 4 6 8 10 12 14 16 Canadian Freeze-Thaw , % Loss

Figure R.31: Specific Gravity (Bulk) vs. Canadian Freeze-Thaw

3 2.9 Specific Gravity (SSD) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 0 2 4 6 8 10 12 14 16 Canadian Freeze-Thaw , % Loss

y = -0.0214Ln(x) + 2.6781 R2 = 0.0188

Figure R.32: Specific Gravity (SSD) vs. Canadian Freeze-Thaw

531

3 2.9 Specific Gravity (Apparent) 2.8 2.7 2.6 2.5 2.4 2.3 0 2 4 6 8 10 12 14 16 Canadian Freeze-Thaw , % Loss

y = 0.0006x 2 - 0.0087x + 2.7313 R2 = 0.0094

Figure R.33: Specific Gravity (Apparent) vs. Canadian Freeze-Thaw

4.5 4 3.5 Particle Shape Factor 3 2.5 2 1.5 1 0.5 0 0 2 4 6 8 10 12 14 16 Canadian Freeze-Thaw , % Loss

y = -0.0017x 2 + 0.009x + 2.299 R2 = 0.0066

Figure R.34: Particle Shape Factor vs. Canadian Freeze-Thaw

532

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = 0.9048x + 2.5113 R2 = 0.7476

Figure R.35: Aggregate Crushing Value (SSD) vs. Aggregate Crushing Value

6 5 4 Absorption 3 2 1 0 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = 0.0041x 2 - 0.169x + 3.075 R2 = 0.0111

Figure R.36: Absorption vs. Aggregate Crushing Value

533

3.1 3 2.9 Specific Gravity (Bulk) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = -0.0069x + 2.762 R2 = 0.0679

Figure R.37: Specific Gravity (Bulk) vs. Aggregate Crushing Value

3 2.9 Specific Gravity (SSD) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = -0.0062x + 2.7834 R2 = 0.0953

Figure R.38: Specific Gravity (SSD) vs. Aggregate Crushing Value

534

3 2.9 Specific Gravity (Apparent) 2.8 2.7 2.6 2.5 2.4 2.3 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = -0.0046x + 2.8085 R2 = 0.0905

Figure R.39: Specific Gravity (Apparent) vs. Aggregate Crushing Value

4.5 4 3.5 Particle Shape Factor 3 2.5 2 1.5 1 0.5 0 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = 0.0011x 2 - 0.0553x + 2.9319 R2 = 0.014

Figure R.40: Particle Shape Factor vs. Aggregate Crushing Value

535

6 5 4 Absorption

y = 0.0045x 2 - 0.2027x + 3.5635 R2 = 0.0358

3 2 1 0 0 5 10 15 20 25 30 35 Aggregate Crushing Value (SSD), % Loss

Figure R.41: Absorption vs. Aggregate Crushing Value (SSD)

3.1 3 2.9 Specific Gravity (Bulk) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 0 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

y = -0.0012x 2 + 0.051x + 2.1212 R2 = 0.1477

Figure R.42: Specific Gravity (Bulk) vs. Aggregate Crushing Value (SSD)

536

3 2.9 Specific Gravity (SSD) 2.8 2.7 2.6 2.5 2.4 2.3 0 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

y = -0.0009x 2 + 0.0348x + 2.3232 R2 = 0.1161

Figure R.43: Specific Gravity (SSD) vs. Aggregate Crushing Value (SSD)

3 2.95 Specific Gravity (Apparent) 2.9 2.85 2.8 2.75 2.7 2.65 2.6 2.55 2.5 0 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

y = -0.0006x 2 + 0.0258x + 2.471 R2 = 0.1198

Figure R.44: Specific Gravity (Apparent) vs. Aggregate Crushing Value (SSD)

537

4.5 4 Particle Shape Factor 3.5 3 2.5 2 1.5 1 0 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

y = 0.0011x 2 - 0.0536x + 2.9081 R2 = 0.021

Figure R.45: Particle Shape Factor vs. Aggregate Crushing Value (SSD)

3.1 3 2.9 Specific Gravity (Bulk) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 0 1 2 3 Absorption

y = 2.7503e-0.0359x R2 = 0.6225

Figure R.46: Specific Gravity (Bulk) vs. Absorption

538

3 2.9 Specific Gravity (SSD) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 0 1 2 3 Absorption 4 5 6

y = 2.7438e-0.0242x R2 = 0.5027

Figure R.47: Specific Gravity (SSD) vs. Absorption

3 2.9 Specific Gravity (Apparent) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 0 1 2 3 Absorption 4 5 6

y = 2.7475e-0.0097x R2 = 0.1472

Figure R.48: Specific Gravity (Apparent) vs. Absorption

539

4.5 4 3.5 Particle Shape Factor 3 2.5 2 1.5 1 0.5 0 0 1 2 3 Absorption 4 5 6

y = 0.0864x 2 - 0.3741x + 2.5291 R2 = 0.0639

Figure R.49: Particle Shape Factor vs. Absorption

3.2 3 Specific Gravity (SSD) 2.8 2.6 2.4 2.2 2 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3 3.1 Specific Gravity (Bulk)

y = 0.2846x 2 - 0.6547x + 2.4128 R2 = 0.9777

Figure R.50: Specific Gravity (SSD) vs. Specific Gravity (Bulk)

540

3.2 3 Specific Gravity (Apparent) 2.8 2.6 2.4 2.2 2 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3 3.1 Specific Gravity (Bulk)

y = 0.6543x 2 - 2.863x + 5.714 R2 = 0.7224

Figure R.51: Specific Gravity (Apparent) vs. Specific Gravity (Bulk)

4.5 4 3.5 Particle Shape Factor 3 2.5 2 1.5 1 0.5 0 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3 3.1 Specific Gravity (Bulk)

y = 2.9614x 2 - 15.018x + 21.253 R2 = 0.0411

Figure R.52: Particle Shape Factor vs. Specific Gravity (Bulk)

541

3.2 3 Specific Gravity (Apparent) 2.8 2.6 2.4 2.2 2 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3 3.1 3.2 Specific Gravity (SSD)

y = 0.6519x 2 - 2.7522x + 5.4202 R2 = 0.8313

Figure R.53: Specific Gravity (Apparent) vs. Specific Gravity (SSD)

4.5 4 3.5 Particle Shape Factor 3 2.5 2 1.5 1 0.5 0 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3 3.1 3.2 Specific Gravity (SSD)

y = 4.6819x 2 - 24.215x + 33.524 R2 = 0.0398

Figure R.54: Particle Shape Factor vs. Specific Gravity (SSD)

542

4.5 4 3.5 Particle Shape Factor 3 2.5 2 1.5 1 0.5 0 2.45

y = 6.1296x 2 - 32.288x + 44.737 R2 = 0.0433

2.5

2.55

2.6

2.65

2.7

2.75

2.8

2.85

2.9

2.95

Specific Gravity (Apparent)

Figure R.55: Particle Shape Factor vs. Specific Gravity (Apparent)

543

Appendix S:

Test Correlation Graphs for the Partial Data Set II

545

40 Magnesium Sulfate Soundness, % Loss 35 30 25 20 15 10 5 0 -5 0 5 10 15 20 25 30 35

y = 0.7627x - 2.5487 R2 = 0.4523

Micro-Deval, % Loss

Figure S.1:

Magnesium Sulfate Soundness vs. Micro-Deval

50 45 40 LA Abrasion, % Loss 35 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 Micro-Deval, % Loss

y = 15.329x 0.1693 R2 = 0.1072

Figure S.2:

L.A. Abrasion vs. Micro-Deval

546

12 10

Canadian Freeze-Thaw, % Loss

y = 1.7951e0.0442x R2 = 0.2317

8 6 4

2 0 0 5 10 15 20 25 30 35 Micro-Deval, % Loss

Figure S.3:

Canadian Freeze-Thaw vs. Micro-Deval

35 Aggregate Crushing Value, % Loss 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 Micro-Deval, % Loss

y = 13.055x 0.1714 R2 = 0.1712

Figure S.4:

Aggregate Crushing Value vs. Micro-Deval

547

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 Micro-Deval, % Loss

y = 13.409x 0.1752 R2 = 0.1898

Figure S.5:

Aggregate Crushing Value (SSD) vs. Micro-Deval

4 3.5 3 Absorption 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 Micro-Deval, % Loss

y = 0.0035x 2 - 0.0459x + 1.0546 R2 = 0.2781

Figure S.6:

Absorption vs. Micro-Deval

548

3.5 3 Specific Gravity (Bulk) 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 Micro-Deval, % Loss

y = -0.0006x 2 + 0.0156x + 2.5595 R2 = 0.1662

Figure S.7:

Specific Gravity (Bulk) vs. Micro-Deval

2.95 2.9 2.85 Specific Gravity (SSD) 2.8 2.75 2.7 2.65 2.6 2.55 2.5 2.45 2.4 0 5 10 15

y = -0.0005x 2 + 0.0137x + 2.5914 R2 = 0.174

20

25

30

35

Micro-Deval, % Loss

Figure S.8:

Specific Gravity (SSD) vs. Micro-Deval

549

2.95 2.9 Specific Gravity (Apparent) 2.85 2.8 2.75 2.7 2.65 2.6 2.55 2.5 0 5 10 15 20 25 30 35 Micro-Deval, % Loss

y = -0.0004x 2 + 0.0128x + 2.6316 R2 = 0.1232

Figure S.9:

Specific Gravity (Apparent) vs. Micro-Deval

4 3.5 3 Particle Shape Factor 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 Micro-Deval, % Loss

y = 0.0715Ln(x) + 2.0514 R2 = 0.0098

Figure S.10: Particle Shape Factor vs. Micro-Deval

550

50 45 40 LA Abrasion, % Loss 35 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 40 Magnesium Sulfate Soundness, % Loss

y = 19.285x 0.1196 R2 = 0.1567

Figure S.11: L.A. Abrasion vs. Magnesium Sulfate Soundness

12 10 8

Canadian Freeze-Thaw, % Loss

y = 1.8811x 0.3483 R2 = 0.3201

6 4 2 0 0 5 10 15 20 25 30 35 40 Magnesium Sulfate Soundness, % Loss

Figure S.12: Canadian Freeze-Thaw vs. Magnesium Sulfate Soundness

551

35 Aggregate Crushing Value, % Loss 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 40 Magnesium Sulfate Soundness, % Loss

y = 17.984x 0.0637 R2 = 0.0701

Figure S.13: Aggregate Crushing Value vs. Magnesium Sulfate Soundness

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 40 Magnesium Sulfate Soundness, % Loss

y = 19.02x 0.0506 R2 = 0.047

Figure S.14: Aggregate Crushing Value (SSD) vs. Magnesium Sulfate Soundness

552

4 3.5 3 Absorption 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 Magnesium Sulfate Soundness, % Loss

y = 0.4266x 0.4976 R2 = 0.4692

Figure S.15: Absorption vs. Magnesium Sulfate Soundness

3.5 3 Specific Gravity (Bulk) 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 Magnesium Sulfate Soundness, % Loss

y = 2.7062x -0.0168 R2 = 0.2115

Figure S.16: Specific Gravity (Bulk) vs. Magnesium Sulfate Soundness

553

2.95 2.9 2.85 Specific Gravity (SSD) 2.8 2.75 2.7 2.65 2.6 2.55 2.5 2.45 2.4 0 5 10 15 20 25 30 35 40 Magnesium Sulfate Soundness, % Loss

y = 2.7126x -0.0123 R2 = 0.1703

Figure S.17: Specific Gravity (SSD) vs. Magnesium Sulfate Soundness

3 2.95 2.9 Specific Gravity (Apparent) 2.85 2.8 2.75 2.7 2.65 2.6 2.55 2.5 2.45 0 5 10 15 20 25 30

y = 2.7461x -0.0078 R2 = 0.0883

35

40

45

Magnesium Sulfate Soundness, % Loss

Figure S.18: Specific Gravity (Apparent) vs. Magnesium Sulfate Soundness

554

4 3.5 3 Particle Shape Factor 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 Magnesium Sulfate Soundness, % Loss

y = 0.001x 2 - 0.0251x + 2.3221 R2 = 0.0326

Figure S.19: Particle Shape Factor vs. Magnesium Sulfate Soundness

12

Canadian Freeze-Thaw, % Loss

y = -0.0079x 2 + 0.4147x - 0.783 R2 = 0.0681 10

8 6 4 2 0 0 5 10 15 20 25 30 35 40 45 50 LA Abrasion, % Loss

Figure S.20: Canadian Freeze-Thaw vs. L.A. Abrasion

555

35 Aggregate Crushing Value, % Loss 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 40 45 50 LA Abrasion, % Loss

y = 2.6341x 0.6367 R2 = 0.7013

Figure S.21: Aggregate Crushing Value vs. L.A. Abrasion

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0 5 10 15 20 25 30 35 40 45 50 LA Abrasion, % Loss

y = 0.4828x + 8.982 R2 = 0.6132

Figure S.22: Aggregate Crushing Value (SSD) vs. L.A. Abrasion

556

4 3.5 3 Absorption 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 45 50 LA Abrasion, % Loss

y = 0.5819Ln(x) - 0.5564 R2 = 0.0406

Figure S.23: Absorption vs. L.A. Abrasion

3 2.9 2.8 Specific Gravity (Bulk) 2.7 2.6 2.5 2.4 2.3 2.2 2.1 2 0 5 10 15 20 25 30 35 40 45 50 LA Abrasion, % Loss

y = -0.1168Ln(x) + 2.9974 R2 = 0.1262

Figure S.24: Specific Gravity (Bulk) vs. L.A. Abrasion

557

3 2.9 Specific Gravity (SSD) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 0 5 10 15 20 25 30 35 40 45 50 LA Abrasion, % Loss

y = -0.0813Ln(x) + 2.9132 R2 = 0.089

Figure S.25: Specific Gravity (SSD) vs. L.A. Abrasion

3 2.9 Specific Gravity (Apparent) 2.8 2.7 2.6 2.5 2.4 2.3 0 5 10 15 20 25 30 35 40 45 50 LA Abrasion, % Loss

y = 2.9317x -0.0249 R2 = 0.0898

Figure S.26: Specific Gravity (Apparent) vs. L.A. Abrasion

558

4 3.5 3 Particle Shape Factor 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 45 50 LA Abrasion, % Loss

y = 0.0006x 2 - 0.0354x + 2.7154 R2 = 0.0223

Figure S.27: Particle Shape Factor vs. L.A. Abrasion

50 Aggregate Crushing Value, % Loss 45 40 35 30 25 20 15 10 0 2 4 6 8 10 12 Canadian Freeze-Thaw , % Loss

y = 0.0891x 2 - 0.8859x + 22.217 R2 = 0.0174

Figure S.28: Aggregate Crushing Value vs. Canadian Freeze-Thaw

559

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 0 2 4 6 8 10 12 Canadian Freeze-Thaw , % Loss

y = 0.1077x 2 - 1.1763x + 23.628 R2 = 0.0249

Figure S.29: Aggregate Crushing Value (SSD) vs. Canadian Freeze-Thaw

4 3.5 3 Absorption 2.5 2 1.5 1 0.5 0 0 2 4 6 8 10 12 Canadian Freeze-Thaw , % Loss

y = 0.593x 0.4155 R2 = 0.1187

Figure S.30: Absorption vs. Canadian Freeze-Thaw

560

3 2.9 2.8 Specific Gravity (Bulk) 2.7 2.6 2.5 2.4 2.3 2.2 2.1 2 0 2 4 6 8 10 12 Canadian Freeze-Thaw , % Loss

y = 2.6581x -0.0098 R2 = 0.0245

Figure S.31: Specific Gravity (Bulk) vs. Canadian Freeze-Thaw

3 2.9 Specific Gravity (SSD) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 0 2 4 6 8 10 12 Canadian Freeze-Thaw , % Loss

y = 2.6779e-0.0023x R2 = 0.0335

Figure S.32: Specific Gravity (SSD) vs. Canadian Freeze-Thaw

561

2.95 2.9 Specific Gravity (Apparent) 2.85 2.8 2.75 2.7 2.65 2.6 2.55 2.5 0 2 4 6 8 10 12 Canadian Freeze-Thaw , % Loss

y = -4E-05x 2 - 0.0009x + 2.7103 R2 = 0.0023

Figure S.33: Specific Gravity (Apparent) vs. Canadian Freeze-Thaw

4 3.5 3 Particle Shape Factor 2.5 2 1.5 1 0.5 0 0 2 4 6 8 10 12 Canadian Freeze-Thaw , % Loss

y = -2E-05x 2 - 0.0079x + 2.2788 R2 = 0.0021

Figure S.34: Particle Shape Factor vs. Canadian Freeze-Thaw

562

35 Aggregate Crushing Value (SSD), % Loss 30 25 20 15 10 5 0 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = 0.9417x + 1.9222 R2 = 0.8567

Figure S.35: Aggregate Crushing Value (SSD) vs. Aggregate Crushing Value

4 3.5 3 Absorption 2.5 2 1.5 1 0.5 0 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = 0.0007x 2 - 0.0158x + 1.2877 R2 = 0.0053

Figure S.36: Absorption vs. Aggregate Crushing Value

563

3 2.9 Specific Gravity (Bulk) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = -0.0043x + 2.715 R2 = 0.0395

Figure S.37: Specific Gravity (Bulk) vs. Aggregate Crushing Value

3 2.9 Specific Gravity (SSD) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = -6E-05x 2 - 0.0009x + 2.6977 R2 = 0.0348

Figure S.38: Specific Gravity (SSD) vs. Aggregate Crushing Value

564

2.95 2.9 Specific Gravity (Apparent) 2.85 2.8 2.75 2.7 2.65 2.6 2.55 2.5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

y = -0.0006x 2 + 0.0214x + 2.5223 R2 = 0.0843

Figure S.39: Specific Gravity (Apparent) vs. Aggregate Crushing Value

4 3.5 3 2.5 2 1.5 1 0.5 0 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value, % Loss

Particle Shape Factor

y = 2.5625x -0.0531 R2 = 0.0049

Figure S.40: Particle Shape Factor vs. Aggregate Crushing Value

565

4 3.5 3 Absorption 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

y = 0.0083x + 1.0871 R2 = 0.002

Figure S.41: Absorption vs. Aggregate Crushing Value (SSD)

3 2.9 Specific Gravity (Bulk) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 0 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

y = -0.0003x 2 + 0.0113x + 2.5348 R2 = 0.0197

Figure S.42: Specific Gravity (Bulk) vs. Aggregate Crushing Value (SSD)

566

3 2.9 Specific Gravity (SSD) 2.8 2.7 2.6 2.5 2.4 2.3 0 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

y = -0.0003x 2 + 0.0098x + 2.5703 R2 = 0.0186

Figure S.43: Specific Gravity (SSD) vs. Aggregate Crushing Value (SSD)

2.95 2.9 Specific Gravity (Apparent) 2.85 2.8 2.75 2.7 2.65 2.6 2.55 2.5 0 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

y = -0.0006x 2 + 0.026x + 2.4541 R2 = 0.103

Figure S.44: Specific Gravity (Apparent) vs. Aggregate Crushing Value (SSD)

567

3.5 Particle Shape Factor 3

y = 0.0016x 2 - 0.0785x + 3.1168 R2 = 0.0312

2.5

2 1.5

1 0 5 10 15 20 25 30 35 40 45 50 Aggregate Crushing Value (SSD), % Loss

Figure S.45: Particle Shape Factor vs. Aggregate Crushing Value (SSD)

3 2.9 Specific Gravity (Bulk) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 0 0.5 1 1.5 2 Absorption 2.5 3 3.5 4

y = 2.7378e-0.0324x R2 = 0.5553

Figure S.46: Specific Gravity (Bulk) vs. Absorption

568

3 2.9 Specific Gravity (SSD) 2.8 2.7 2.6 2.5 2.4 2.3 2.2 0 0.5 1 1.5 2 Absorption 2.5 3 3.5 4

y = 2.735e-0.0233x R2 = 0.4649

Figure S.47: Specific Gravity (SSD) vs. Absorption

2.95 2.9 Specific Gravity (Apparent) 2.85 2.8 2.75 2.7 2.65 2.6 2.55 2.5 0 0.5 1 1.5 2 Absorption 2.5 3 3.5 4

y = 2.7087x -0.0113 R2 = 0.116

Figure S.48: Specific Gravity (Apparent) vs. Absorption

569

4 3.5 3 2.5 2 1.5 1 0.5 0 0 0.5 1 1.5 2 Absorption 2.5 3 3.5 4

y = 0.0763x 2 - 0.351x + 2.4882 R2 = 0.0629

Particle Shape Factor

Figure S.49: Particle Shape Factor vs. Absorption

3 2.9 2.8 Specific Gravity (SSD) 2.7 2.6 2.5 2.4 2.3 2.2 2.1 2 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3 3.1 Specific Gravity (Bulk)

y = 0.2823x 2 - 0.6642x + 2.4531 R2 = 0.9693

Figure S.50: Specific Gravity (SSD) vs. Specific Gravity (Bulk)

570

2.95 2.9 Specific Gravity (Apparent) 2.85 2.8 2.75 2.7 2.65 2.6 2.55 2.5 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3 3.1 Specific Gravity (Bulk)

y = 0.5127x 2 - 2.1431x + 4.8003 R2 = 0.6671

Figure S.51: Specific Gravity (Apparent) vs. Specific Gravity (Bulk)

3.6 3.4 3.2 3 2.8 2.6 2.4 2.2 2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3 3.1 Specific Gravity (Bulk)

y = 5.4565x 2 - 28.311x + 38.883 R2 = 0.0636

Particle Shape Factor

Figure S.52: Particle Shape Factor vs. Specific Gravity (Bulk)

571

2.95 2.9 Specific Gravity (Apparent) 2.85 2.8 2.75 2.7 2.65 2.6 2.55 2.5 2.4 2.45 2.5 2.55 2.6 2.65 2.7 2.75 2.8 2.85 2.9 2.95 Specific Gravity (SSD)

y = 0.438x 2 - 1.6274x + 3.9424 R2 = 0.795

Figure S.53: Specific Gravity (Apparent) vs. Specific Gravity (SSD)

4 3.5 3 2.5 2 1.5 1 0.5 0 2.4 2.5 2.6 2.7 2.8 2.9 Specific Gravity (SSD)

Particle Shape Factor

y = 7.9849x 2 - 41.746x + 56.719 R2 = 0.0667

Figure S.54: Particle Shape Factor vs. Specific Gravity (SSD)

572

4 3.5 3 2.5 2 1.5 1 0.5 0 2.5 2.55 2.6 2.65 2.7 2.75 2.8 2.85 2.9 2.95 Specific Gravity (Apparent)

Particle Shape Factor

y = 10.873x 2 - 58.205x + 80.066 R2 = 0.0544

Figure S.55: Particle Shape Factor vs. Specific Gravity (Apparent)

573

Appendix T:

Test Correlation Tables for Full Data Set, Partial Data Set I, and Partial Data Set II

575

ID # 1 2 3 4 5 6 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

MD 17.0 19.2 29.5 22.2 9.7 11.0 13.1 27.6 4.2 7.1 10.3 12.6 8.8 8.6 9.1 16.9 27.0 7.3 13.7 9.0 9.5 6.4 7.9 48.8 19.6 14.8 3.1 21.5 21.4 7.8 12.5 11.7 15.6 19.3 22.5 10.5 23.0 19.8 19.0 4.4 13.7 21.9 9.8 13.1 14.8 16.9

MSS 12.2 68.9 70.3 23.9 1.1 3.0 2.7 15.6 2.6 3.6 5.9 10.0 2.1 1.1 1.8 3.9 12.4 0.4 4.6 3.7 6.1 1.4 1.5 69.4 12.9 0.7 6.0 10.5 9.1 4.8 1.9 1.9 5.1 8.1 3.4 0.6 9.0 22.0 0.3 6.4 3.8 6.9 8.3 10.0 8.7

LAA 31 66 57 29 21 26 14 13 21 14 15 31 18 17 27 14 32 15 59 62 57 37 52 34 24 35 15 27 31 19 24 21 13 42 34 45 16 23 35 18 25 29 20 25 24 25

CFT 7.0 3.1 4.9 5.8 2.4 2.8 2.2 9.3 2.5 3.5 6.0 3.5 2.6 2.0 2.0 7.1 6.6 2.1 2.4 2.3 2.0 1.1 1.6 22.4 5.0 3.0 2.3 2.3 1.5 5.7 2.4 1.6 5.8 3.4 2.6 1.7 1.3 4.6 3.2 1.6 9.0 2.5 10.4 6.8 7.5 9.1

ACV 20 27 25 22 20 21 14 16 15 12 14 24 18 15 23 16 29 13 28 32 32 24 26 26 46 26 15 30 26 16 19 18 13 26 23 30 16 20 25 17 23 25 20 23 23 22

WCV 19.00 30.00 26.00 24.50 19.65 24.00 16.00 19.00 17.00 14.00 15.80 24.20 19.80 16.83 23.00 16.70 28.00 15.00 31.00 31.25 32.95 26.00 28.00 24.00 50.15 7.60 16.00 27.00 26.65 18.50 20.35 20.00 15.87 25.93 23.00 31.15 17.30 18.00 23.00 17.10 21.93 24.40 22.80 21.00 24.80 20.65

ABS 1.6 2.4 2.9 2.2 0.6 0.7 1.3 2.2 0.8 1.1 0.7 0.8 0.5 0.3 0.5 0.9 3.4 0.3 0.8 0.7 0.5 0.5 0.5 5.0 1.0 0.6 3.3 3.8 1.5 1.8 0.3 0.4 0.3 0.7 0.7 0.5 1.1 2.7 0.3 0.8 0.9 1.3 0.8 1.0 1.0

SGBULK 2.58 2.44 2.43 2.51 2.75 2.67 2.76 2.83 2.57 2.70 2.72 2.59 2.76 2.76 2.63 2.59 2.36 2.82 2.63 2.61 2.65 2.61 2.63 2.45 2.60 2.62 2.32 2.42 2.71 2.59 3.01 2.66 2.88 2.67 2.61 2.90 2.66 2.53 2.73 2.65 2.59 2.69 2.67 2.64 2.64

SGSSD 2.63 2.49 2.50 2.57 2.76 2.69 2.79 2.89 2.60 2.73 2.74 2.62 2.77 2.77 2.65 2.61 2.44 2.83 2.65 2.63 2.66 2.63 2.64 2.57 2.62 2.63 2.40 2.52 2.75 2.64 3.02 2.67 2.88 2.69 2.62 2.91 2.69 2.60 2.74 2.67 2.61 2.73 2.69 2.67 2.66

SGAPP 2.70 2.58 2.62 2.66 2.79 2.72 2.86 3.01 2.63 2.78 2.77 2.65 2.80 2.79 2.67 2.65 2.56 2.84 2.69 2.66 2.68 2.65 2.66 2.79 2.67 2.66 2.52 2.67 2.82 2.72 3.05 2.69 2.90 2.72 2.65 2.94 2.74 2.72 2.75 2.71 2.66 2.79 2.72 2.71 2.74

PSF 1.66 1.78 1.57 2.17 2.29 3.06 2.09 1.76 1.53 1.86 2.55 2.22 3.07 2.53 1.53 2.86 1.81 2.33 2.91 3.10 1.90 2.17 1.54 2.91 3.19 2.71 1.80 2.10 3.03 2.36 2.05 2.36 2.90 2.75 1.63 2.46 3.19 3.71 1.46 2.38 2.53 2.66 2.57 1.53 2.17 2.22

576

50 51 52 53 54 55 56 57 58 59 60 61 62 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 94 95 96 97 98 99

21.6 1.6 1.4 16.4 7.3 17.3 20.6 34.3 4.0 24.2 3.7 8.3 26.9 8.0 21.0 23.1 9.9 5.5 10.3 13.6 12.3 13.9 7.4 9.4 21.8 30.9 12.1 18.2 14.4 27.4 17.7 20.4 9.0 18.6 19.1 7.7 8.7 5.4 13.9 28.9 12.5 10.0 19.6 9.9 17.3 27.8 10.9 10.7

23.8 6.6 3.5 4.9 1.5 2.7 1.3 20.0 3.2 36.7 2.1 9.7 28.3 7.4 14.8 7.4 4.2 0.7 3.4 4.5 1.2 12.1 1.0 1.1 6.0 39.0 5.9 16.9 5.5 17.7 20.8 12.6 6.7 11.9 16.3 1.3 2.0 1.0 6.9 39.1 15.7 2.8 18.8 6.4 18.7 31.2 6.3 8.3

29 16 16 16 36 17 21 45 18 27 15 22 32 43 22 31 22 24 12 11 23 16 14 19 17 26 18 25 22 27 27 32 22 26 24 33 38 14 23 36 37 26 32 22 26 18 24 31

8.8 3.9 3.6 6.0 1.0 1.3 2.4 3.1 3.5 14.8 2.9 4.0 8.8 2.9 6.3 6.4 2.4 5.8 2.9 3.3 3.8 3.5 1.5 3.7 3.5 6.6 8.4 12.1 10.3 8.2 3.6 4.3 2.5 4.4 10.6 1.7 1.3 1.4 4.5 10.1 4.2 3.4 12.9 2.7 8.4 13.7 2.9 4.8

22 11 13 16 27 16 19 41 13 21 15 16 24 31 21 30 22 19 13 15 21 13 15 20 17 21 20 25 22 19 21 22 12 19 23 28 28 13 20 24 25 26 27 19 22 16 23 23

20.00 10.00 12.00 19.80 27.80 18.65 23.35 44.40 13.00 20.00 13.00 18.00 23.00 32.30 19.00 32.50 22.50 16.90 15.00 14.00 20.05 15.00 17.00 18.00 15.00 22.80 20.00 22.00 24.85 18.00 25.00 22.00 15.00 24.00 22.40 27.40 28.95 15.05 20.95 24.15 24.70 25.80 30.30 19.60 22.45 20.35 26.45 27.00

2.2 3.1 1.8 0.7 0.4 0.2 0.4 4.5 1.0 4.0 0.8 1.3 2.9 2.2 0.8 5.7 0.7 0.2 1.8 2.1 0.5 1.7 0.9 0.3 1.5 2.9 0.4 0.6 0.6 3.7 3.8 3.7 1.7 1.8 0.6 0.5 0.5 0.1 1.4 2.5 0.9 0.3 0.8 1.3 4.7 1.7 2.3 0.6

2.59 2.40 2.51 2.79 2.62 2.88 2.75 2.27 2.65 2.41 2.60 2.65 2.49 2.63 2.65 2.26 2.63 2.67 2.55 2.66 2.75 2.58 2.88 2.72 2.62 2.49 2.70 2.69 2.68 2.49 2.47 2.47 2.63 2.61 2.67 2.66 2.66 2.73 2.60 2.39 2.63 2.82 2.67 2.64 2.32 2.68 2.52 2.64

2.65 2.48 2.55 2.80 2.63 2.87 2.77 2.37 2.67 2.51 2.62 2.69 2.56 2.69 2.67 2.39 2.65 2.68 2.60 2.71 2.77 2.62 2.90 2.72 2.66 2.57 2.71 2.70 2.70 2.58 2.57 2.56 2.68 2.65 2.68 2.67 2.67 2.74 2.64 2.45 2.66 2.83 2.69 2.67 2.43 2.73 2.58 2.66

2.75 2.60 2.62 2.84 2.65 2.86 2.79 2.53 2.72 2.67 2.66 2.75 2.68 2.79 2.71 2.60 2.68 2.69 2.67 2.81 2.79 2.70 2.96 2.74 2.73 2.69 2.73 2.73 2.73 2.75 2.73 2.72 2.76 2.73 2.71 2.69 2.69 2.74 2.70 2.54 2.70 2.84 2.72 2.74 2.61 2.81 2.68 2.69

2.00 2.43 2.12 3.22 2.01 2.75 2.89 2.32 1.98 1.81 2.27 1.92 2.40 2.16 2.02 1.77 2.24 2.12 2.27 2.13 2.15 1.89 2.37 1.95 2.17 2.30 1.52 1.87 2.06 1.88 2.46 2.00 1.66 2.07 2.50 2.21 2.08 2.53 2.36 1.84 3.82 1.94 2.21 1.76 3.00 2.74 2.50 1.86

577

100 101 102 103 104 105 106 108 109 110 111 112 113 114 115 116 117

4.4 19.6 39.3 22.6 3.1 10.6 14.2 20.3 12.4 16.7 6.7 22.4 7.1 8.6 14.5 9.7 30.4

0.8 5.4 30.9 22.0 1.8 9.4 13.8 13.5 3.5 8.3 3.5 5.0 1.9 5.9 1.9 9.2 33.0

23 52 56 29 18 34 27 34 23 30 20 28 34 28 22 25 35

1.1 1.9 3.6 10.8 0.6 1.1 3.3 2.0 4.3 2.8 0.9 2.9 1.3 2.5 4.7 3.6 5.6

18 26 48 24 17 21 17 18 22 22 13 20 22 23 20 17 23

21.00 32.00 45.00 23.40 15.00 21.00 17.30 20.90 20.00 14.00 19.00 25.00 25.00 19.00 15.00 21.00

0.7 0.7 4.7 2.2 0.3 2.0 2.0 2.4 0.6 1.6 1.6 0.7 0.7 1.0 1.3 1.7 1.5

2.61 2.74 2.29 2.54 2.63 2.57 2.51 2.21 2.67 2.64 2.51 2.72 2.62 2.60 2.62 2.53 2.56

2.63 2.76 2.40 2.59 2.64 2.60 2.56 2.26 2.69 2.69 2.55 2.74 2.64 2.62 2.65 2.57 2.60

2.66 2.79 2.56 2.69 2.65 2.65 2.65 2.33 2.72 2.76 2.62 2.77 2.67 2.67 2.71 2.64 2.67

2.03 2.49 3.41 4.27 1.84 1.95 5.19 2.05 2.14 1.74 2.32 3.91 2.12 2.06 1.90 3.25 3.03

Table T.1:

Full Data Set

578

ID # 1 2 3 4 5 6 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

MD 17 19. 2 29. 5 22. 2 9.7 11 13. 1 27. 6 4.2 7.1 10. 3 12. 6 8.8 8.6 9.1 16. 9 27 7.3 13. 7 9 9.5 6.4 7.9 19. 6 14. 8 3.1 21. 5 21. 4 7.8 12. 5 11.

MSS 12.2

LAA 31

CFT 7 3.1

ACV 20 27 25 22 20 21 14 16 15 12 14 24 18 15 23 16 29 13 28 32 32 24 26 26

WCV 19 30 26 24.5 19.65 24 16 19 17 14 15.8 24.2 19.8 16.83 3 23 16.7 28 15 31 31.25 32.95 26 28 24

ABS 1.6 2.4 2.9 2.2 0.6 0.7 1.3 2.2 0.8 1.1 0.7 0.8 0.5 0.3 0.5 0.9 3.4 0.3 0.8 0.7 0.5 0.5 0.5 5

SGBULK 2.58 2.44 2.43 2.51 2.75 2.67 2.76 2.83 2.57 2.7 2.72 2.59 2.76 2.76 2.63 2.59 2.36 2.82 2.63 2.61 2.65 2.61 2.63 2.45

SGSSD 2.63 2.49 2.5 2.57 2.76 2.69 2.79 2.89 2.6 2.73 2.74 2.62 2.77 2.77 2.65 2.61 2.44 2.83 2.65 2.63 2.66 2.63 2.64 2.57

SGAPP 2.7 2.58 2.62 2.66 2.79 2.72 2.86 2.63 2.78 2.77 2.65 2.8 2.79 2.67 2.65 2.56 2.84 2.69 2.66 2.68 2.65 2.66 2.79

PSF 1.66 1.78 1.57 2.17 2.29 3.06 2.09 1.76 1.53 1.86 2.55 2.22 3.07 2.53 1.53 2.86 1.81 2.33 2.91 3.1 1.9 2.17 1.54 2.91 3.19

57 23.9 1.1 3 2.7 15.6 2.6 3.6 5.9 10 2.1 1.1 1.8 3.9 12.4 0.4 4.6 3.7 6.1 1.4 1.5 29 21 26 14 13 21 14 15 31 18 17 27 14 32 15 59 57 37 52 34 24 12.9 0.7 6 10.5 9.1 4.8 1.9 35 15 27 31 19 24 21

4.9 5.75 2.37 5 2.8 2.2 9.3 2.5 3.5 6 3.45 2.55 1.95 2 7.1 6.6 2.1 2.4 2.25 2 1.1 1.6 4.95 3 2.3 2.3 1.45 5.65 2.4 1.6

26 15 30 26 16 19 18

7.6 16 27 26.65 18.5 20.35 20

1 0.6 3.3 3.8 1.5 1.8 0.3

2.6 2.62 2.32 2.42 2.71 2.59 3.01

2.62 2.63 2.4 2.52 2.75 2.64

2.67 2.66 2.52 2.67 2.82 2.72

2.71 1.8 2.1 3.03 2.36 2.05 2.36

579

36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 64 65 66 67 68

7 15. 6 19. 3 22. 5 10. 5 23 19. 8 19 4.4 13. 7 21. 9 9.8 13. 1 14. 8 16. 9 21. 6 1.6 1.4 16. 4 7.3 17. 3 20. 6 34. 3 4 24. 2 3.7 8.3 26. 9 8 21 23. 1 9.9 5.5

1.9 5.1 8.1 3.4 0.6 9 22 0.3 6.4 3.8 6.9 8.3 10 8.7 23.8 6.6 3.5 4.9 1.5 2.7 1.3 20 3.2 36.7 2.1 9.7 28.3 7.4 14.8 7.4 4.2 0.7

13 42 34 45 16 23 35 18 25 29 20 25 24 25 29 16 16 16 36 17 21 45 18 27 15 22 32 43 22 31 22 24

5.8 3.4 2.6 1.65 1.25 4.6 3.2 1.55 8.95 2.5 10.4 6.8 7.5 9.05 8.8 3.9 3.6 6 1 1.3 2.35 3.05 3.5 14.8 2.9 4 8.8 2.85 6.3 6.35 2.35 5.75

13 26 23 30 16 20 25 17 23 25 20 23 23 22 22 11 13 16 27 16 19 13 21 15 16 24 31 21 30 22 19

15.86 7 25.93 3 23 31.15 17.3 18 23 17.1 21.93 3 24.4 22.8 21 24.8 20.65 20 10 12 19.8 27.8 18.65 23.35 13 20 13 18 23 32.3 19 32.5 22.5 16.9

0.4 0.3 0.7 0.7 0.5 1.1 2.7 0.3 0.8 0.9 1.3 0.8 1 1 2.2 3.1 1.8 0.7 0.4 0.2 0.4 4.5 1 4 0.8 1.3 2.9 2.2 0.8 0.7 0.2

2.66 2.88 2.67 2.61 2.9 2.66 2.53 2.73 2.65 2.59 2.69 2.67 2.64 2.64 2.59 2.4 2.51 2.79 2.62 2.88 2.75 2.27 2.65 2.41 2.6 2.65 2.49 2.63 2.65 2.26 2.63 2.67

2.67 2.88 2.69 2.62 2.91 2.69 2.6 2.74 2.67 2.61 2.73 2.69 2.67 2.66 2.65 2.48 2.55 2.8 2.63 2.87 2.77 2.37 2.67 2.51 2.62 2.69 2.56 2.69 2.67 2.39 2.65 2.68

2.69 2.9 2.72 2.65 2.94 2.74 2.72 2.75 2.71 2.66 2.79 2.72 2.71 2.74 2.75 2.6 2.62 2.84 2.65 2.86 2.79 2.53 2.72 2.67 2.66 2.75 2.68 2.79 2.71 2.6 2.68 2.69

2.9 2.75 1.63 2.46 3.19 3.71 1.46 2.38 2.53 2.66 2.57 1.53 2.17 2.22 2 2.43 2.12 3.22 2.01 2.75 2.89 2.32 1.98 1.81 2.27 1.92 2.4 2.16 2.02 1.77 2.24 2.12

580

69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 94 95 96 97 98 99

10. 3 13. 6 12. 3 13. 9 7.4 9.4 21. 8 30. 9 12. 1 18. 2 14. 4 27. 4 17. 7 20. 4 9 18. 6 19. 1 7.7 8.7 5.4 13. 9 28. 9 12. 5 10 19. 6 9.9 17. 3 27. 8 10. 9 10.

3.4 4.5 1.2 12.1 1 1.1 6 39 5.9 16.9 5.5 17.7 20.8 12.6 6.7 11.9 16.3 1.3 2 1 6.9 39.1 15.7 2.8 18.8 6.4 18.7 31.2 6.3 8.3

12 11 23 16 14 19 17 26 18 25 22 27 27 32 22 26 24 33 38 14 23 36 37 26 32 22 26 18 24 31

2.9 3.3 3.75 3.5 1.5 3.7 3.5 6.6 8.4 12.1 10.2 5 8.2 3.6 4.3 2.5 4.4 10.5 5 1.7 1.3 1.4 4.5 10.1 4.2 3.35 12.9 2.7 8.4 13.7 2.9 4.8

13 15 21 13 15 20 17 21 20 25 22 19 21 22 12 19 23 28 28 13 20 24 25 26 27 19 22 16 23 23

15 14 20.05 15 17 18 15 22.8 20 22 24.85 18 25 22 15 24 22.4 27.4 28.95 15.05 20.95 24.15 24.7 25.8 30.3 19.6 22.45 20.35 26.45 27

1.8 2.1 0.5 1.7 0.9 0.3 1.5 2.9 0.4 0.6 0.6 3.7 3.8 3.7 1.7 1.8 0.6 0.5 0.5 0.1 1.4 2.5 0.9 0.3 0.8 1.3 4.7 1.7 2.3 0.6

2.55 2.66 2.75 2.58 2.88 2.72 2.62 2.49 2.7 2.69 2.68 2.49 2.47 2.47 2.63 2.61 2.67 2.66 2.66 2.73 2.6 2.39 2.63 2.82 2.67 2.64 2.32 2.68 2.52 2.64

2.6 2.71 2.77 2.62 2.9 2.72 2.66 2.57 2.71 2.7 2.7 2.58 2.57 2.56 2.68 2.65 2.68 2.67 2.67 2.74 2.64 2.45 2.66 2.83 2.69 2.67 2.43 2.73 2.58 2.66

2.67 2.81 2.79 2.7 2.96 2.74 2.73 2.69 2.73 2.73 2.73 2.75 2.73 2.72 2.76 2.73 2.71 2.69 2.69 2.74 2.7 2.54 2.7 2.84 2.72 2.74 2.61 2.81 2.68 2.69

2.27 2.13 2.15 1.89 2.37 1.95 2.17 2.3 1.52 1.87 2.06 1.88 2.46 2 1.66 2.07 2.5 2.21 2.08 2.53 2.36 1.84 3.82 1.94 2.21 1.76 3 2.74 2.5 1.86

581

100 101 102 103 104 105 106 108 109 110 111 112 113 114 115 116 117

7 4.4 19. 6 39. 3 22. 6 3.1 10. 6 14. 2 20. 3 12. 4 16. 7 6.7 22. 4 7.1 8.6 14. 5 9.7 30. 4

0.8 5.4 30.9 22 1.8 9.4 13.8 13.5 3.5 8.3 3.5 5 1.9 5.9 1.9 9.2 33

23 52 56 29 18 34 27 34 23 30 20 28 34 28 22 25 35

1.1 1.9 3.6 10.8 0.6 1.1 3.25 2 4.27 5 2.8 0.9 2.9 1.3 2.5 4.7 3.6 5.6

18 26

21 32

0.7 0.7 4.7

2.61 2.74 2.29 2.54 2.63 2.57 2.51 2.21 2.67 2.64 2.51 2.72 2.62 2.6 2.62 2.53 2.56

2.63 2.76 2.4 2.59 2.64 2.6 2.56

2.66 2.79 2.56 2.69 2.65 2.65 2.65

2.03 2.49 3.41 1.84 1.95

24 17 21 17 18 22 22 13 20 22 23 20 17 23

23.4 15 21 17.3

2.2 0.3 2 2 2.4

2.05 2.69 2.69 2.55 2.74 2.64 2.62 2.65 2.57 2.6 2.72 2.76 2.62 2.77 2.67 2.67 2.71 2.64 2.67 2.14 1.74 2.32 3.91 2.12 2.06 1.9 3.25 3.03

20.9 20 14 19 25 25 19 15 21

0.6 1.6 1.6 0.7 0.7 1 1.3 1.7 1.5

Table T.2:

Partial Data Set I

582

ID # 1 2 3 4 5 6 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

MD 17 19.2 29.5 22.2 9.7 11 13.1 27.6 4.2 7.1 10.3 12.6 8.8 8.6 9.1 16.9 27 7.3 13.7 9 9.5 6.4 7.9 19.6 14.8 3.1 21.5 21.4 7.8 12.5 11.7 15.6 19.3 22.5 10.5 23 19.8 19 4.4 13.7 21.9 9.8 13.1 14.8

MSS 12.2

LAA 31

23.9 1.1 3 2.7 15.6 2.6 3.6 5.9 10 2.1 1.1 1.8 3.9 12.4 0.4 4.6 3.7 6.1 1.4 1.5

29 21 26 14 13 21 14 15 31 18 17 27 14 32 15

37 34 24 35 15 27 31 19 24 21 13 42 34 45 16 23 35 18 25 29 20 25 24

CFT 7 3.1 4.9 5.75 2.375 2.8 2.2 9.3 2.5 3.5 6 3.45 2.55 1.95 2 7.1 6.6 2.1 2.4 2.25 2 1.1 1.6 4.95 3 2.3 2.3 1.45 5.65 2.4 1.6 5.8 3.4 2.6 1.65 1.25 4.6 3.2 1.55 8.95 2.5 10.4 6.8 7.5

ACV 20 27 25 22 20 21 14 16 15 12 14 24 18 15 23 16 29 13 28 32 32 24 26 26 26 15 30 26 16 19 18 13 26 23 30 16 20 25 17 23 25 20 23 23

WCV 19 30 26 24.5 19.65 24 16 19 17 14 15.8 24.2 19.8 16.833 23 16.7 28 15 31 31.25 32.95 26 28 24

ABS 1.6 2.4 2.9 2.2 0.6 0.7 1.3 2.2 0.8 1.1 0.7 0.8 0.5 0.3 0.5 0.9 3.4 0.3 0.8 0.7 0.5 0.5 0.5

SGBULK 2.58 2.44 2.43 2.51 2.75 2.67 2.76 2.83 2.57 2.7 2.72 2.59 2.76 2.76 2.63 2.59 2.36 2.82 2.63 2.61 2.65 2.61 2.63 2.45 2.6 2.62 2.42 2.71 2.59 2.66 2.88 2.67 2.61 2.66 2.53 2.73 2.65 2.59 2.69 2.67 2.64

SGSSD 2.63 2.49 2.5 2.57 2.76 2.69 2.79 2.6 2.73 2.74 2.62 2.77 2.77 2.65 2.61 2.44 2.83 2.65 2.63 2.66 2.63 2.64 2.57 2.62 2.63 2.52 2.75 2.64 2.67 2.88 2.69 2.62 2.69 2.6 2.74 2.67 2.61 2.73 2.69 2.67

SGAPP 2.7 2.58 2.62 2.66 2.79 2.72 2.86 2.63 2.78 2.77 2.65 2.8 2.79 2.67 2.65 2.56 2.84 2.69 2.66 2.68 2.65 2.66 2.79 2.67 2.66 2.52 2.67 2.82 2.72 2.69 2.9 2.72 2.65 2.74 2.72 2.75 2.71 2.66 2.79 2.72 2.71

12.9 0.7 6 10.5 9.1 4.8 1.9 1.9 5.1 8.1 3.4 0.6 9 22 0.3 6.4 3.8 6.9 8.3 10

16 27 26.65 18.5 20.35 20 15.867 25.933 23 31.15 17.3 18 23 17.1 21.933 24.4 22.8 21 24.8

1 0.6 3.3 3.8 1.5 1.8 0.3 0.4 0.3 0.7 0.7 0.5 1.1 2.7 0.3 0.8 0.9 1.3 0.8 1

PSF 1.66 1.78 1.57 2.17 2.29 3.06 2.09 1.76 1.53 1.86 2.55 2.22 3.07 2.53 1.53 2.86 1.81 2.33 2.91 3.1 1.9 2.17 1.54 2.91 3.19 2.71 1.8 2.1 3.03 2.36 2.05 2.36 2.9 2.75 1.63 2.46 3.19 1.46 2.38 2.53 2.66 2.57 1.53 2.17

583

49 50 51 52 53 54 55 56 57 58 59 60 61 62 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 94 95 96 97 98

16.9 21.6 1.6 1.4 16.4 7.3 17.3 20.6 4 24.2 3.7 8.3 26.9 8 21 23.1 9.9 5.5 10.3 13.6 12.3 13.9 7.4 9.4 21.8 30.9 12.1 18.2 14.4 27.4 17.7 20.4 9 18.6 19.1 7.7 8.7 5.4 13.9 28.9 12.5 10 19.6 9.9 17.3 27.8 10.9

8.7 23.8 6.6 3.5 4.9 1.5 2.7 1.3 20 3.2 36.7 2.1 9.7 28.3 7.4 14.8 7.4 4.2 0.7 3.4 4.5 1.2 12.1 1 1.1 6 5.9 16.9 5.5 17.7 20.8 12.6 6.7 11.9 16.3 1.3 2 1 6.9 15.7 2.8 18.8 6.4 18.7 31.2 6.3

25 29 16 16 16 36 17 21 45 18 27 15 22 32 43 22 31 22 24 12 11 23 16 14 19 17 26 18 25 22 27 27 32 22 26 24 33 38 14 23 36 37 26 32 22 26 18 24

9.05 8.8 3.9 3.6 6 1 1.3 2.35 3.05 3.5 2.9 4 8.8 2.85 6.3 6.35 2.35 5.75 2.9 3.3 3.75 3.5 1.5 3.7 3.5 6.6 8.4 10.25 8.2 3.6 4.3 2.5 4.4 10.55 1.7 1.3 1.4 4.5 10.1 4.2 3.35 2.7 8.4 2.9

22 22 11 13 16 27 16 19 13 21 15 16 24 31 21 30 22 19 13 15 21 13 15 20 17 21 20 25 22 19 21 22 12 19 23 28 28 13 20 24 25 26 27 19 22 16 23

20.65 20 10 12 19.8 27.8 18.65 23.35 13 20 13 18 23 32.3 19 32.5 22.5 16.9 15 14 20.05 15 17 18 15 22.8 20 22 24.85 18 25 22 15 24 22.4 27.4 28.95 15.05 20.95 24.15 24.7 25.8 30.3 19.6 22.45 20.35 26.45

1 2.2 3.1 1.8 0.7 0.4 0.2 0.4 1 0.8 1.3 2.9 2.2 0.8 0.7 0.2 1.8 2.1 0.5 1.7 0.9 0.3 1.5 2.9 0.4 0.6 0.6 3.7 3.8 3.7 1.7 1.8 0.6 0.5 0.5 0.1 1.4 2.5 0.9 0.3 0.8 1.3 1.7 2.3

2.64 2.59 2.4 2.51 2.79 2.62 2.88 2.75 2.65 2.41 2.6 2.65 2.49 2.63 2.65 2.63 2.67 2.55 2.66 2.75 2.58 2.88 2.72 2.62 2.49 2.7 2.69 2.68 2.49 2.47 2.47 2.63 2.61 2.67 2.66 2.66 2.73 2.6 2.39 2.63 2.82 2.67 2.64 2.68 2.52

2.66 2.65 2.48 2.55 2.8 2.63 2.87 2.77 2.67 2.51 2.62 2.69 2.56 2.69 2.67 2.65 2.68 2.6 2.71 2.77 2.62 2.72 2.66 2.57 2.71 2.7 2.7 2.58 2.57 2.56 2.68 2.65 2.68 2.67 2.67 2.74 2.64 2.45 2.66 2.83 2.69 2.67 2.43 2.73 2.58

2.74 2.75 2.6 2.62 2.84 2.65 2.86 2.79 2.53 2.72 2.67 2.66 2.75 2.68 2.79 2.71 2.6 2.68 2.69 2.67 2.81 2.79 2.7 2.74 2.73 2.69 2.73 2.73 2.73 2.75 2.73 2.72 2.76 2.73 2.71 2.69 2.69 2.74 2.7 2.54 2.7 2.84 2.72 2.74 2.61 2.81 2.68

2.22 2 2.43 2.12 3.22 2.01 2.75 2.89 2.32 1.98 1.81 2.27 1.92 2.4 2.16 2.02 1.77 2.24 2.12 2.27 2.13 2.15 1.89 2.37 1.95 2.17 2.3 1.52 1.87 2.06 1.88 2.46 2 1.66 2.07 2.5 2.21 2.08 2.53 2.36 1.84 1.94 2.21 1.76 3 2.74 2.5

584

99 100 101 102 103 104 105 106 108 109 110 111 112 113 114 115 116 117

10.7 4.4 19.6 22.6 3.1 10.6 14.2 20.3 12.4 16.7 6.7 22.4 7.1 8.6 14.5 9.7 30.4

8.3 0.8 5.4 30.9 22 1.8 9.4 13.8 13.5 3.5 8.3 3.5 5 1.9 5.9 1.9 9.2 33

31 23

29 18 34 27 34 23 30 20 28 34 28 22 25 35

4.8 1.1 1.9 3.6 10.8 0.6 1.1 3.25 2 4.275 2.8 0.9 2.9 1.3 2.5 4.7 3.6 5.6

23 18 26 24 17 21 17 18 22 22 13 20 22 23 20 17 23

27 21 32 23.4 15 21 17.3 20.9 20 14 19 25 25 19 15 21

0.6 0.7 0.7 2.2 0.3 2 2 2.4 0.6 1.6 1.6 0.7 0.7 1 1.3 1.7 1.5

2.64 2.61 2.74 2.54 2.63 2.57 2.51 2.67 2.64 2.51 2.72 2.62 2.6 2.62 2.53 2.56

2.66 2.63 2.76 2.59 2.64 2.6 2.56 2.69 2.69 2.55 2.74 2.64 2.62 2.65 2.57 2.6

2.69 2.66 2.79 2.56 2.69 2.65 2.65 2.65 2.72 2.76 2.62 2.77 2.67 2.67 2.71 2.64 2.67

1.86 2.03 2.49 3.41 1.84 1.95 2.05 2.14 1.74 2.32 2.12 2.06 1.9 3.25 3.03

Table T.3:

Partial Data Set II

585

Appendix U:

Dr. Eyad Masad AIMS Report

586

Analysis of Shape Characteristics and Abrasion Resistance of University of Texas Aggregates

Report Prepared by

Eyad Masad, Ph.D., P.E

December 17, 2005

587

AGGREGATE IMAGING SYSTEM (AIMS) SHAPE ANALYSIS METHODS

This section discusses the analysis methods used in measuring form, angularity, and texture of aggregates. These methods are part of the AIMS software. AIMS is equipped with top lighting, back lighting, and autofocus microscope. These are necessary elements of the system to capture black and white images, and gray scales images at different magnifications for the analysis of all shape characteristics. The system is computercontrolled to achieve motion in the x, y, and z directions as well as magnification. More details about the design and operations of the system are provided by Masad et al. (2005).
Form

The three dimensions of aggregates are needed to describe the form of coarse aggregates. The camera and microscope assembly are used to capture projections of particles placed on a lighting table. Particle projections are used to measure the longest and shortest dimensions using eigenvector analysis (Masad et al. 2005). In this method, the binary image of an aggregate is treated as a two-dimensional population. Each pixel in the population is treated as a two-dimensional vector. These vectors are used to compute the eigenvectors which are orthogonal to each other. The major and minor axes of an aggregate are aligned along these eigenvectors. A particle depth is measured using the location of the autofocus microscope. First, the microscope is focused automatically on a point on the lighting table. The distance between the lighting table and the lens is recorded as the reference position. Then, the 588

microscope moves in the x and y directions and focuses on a particle surface. The focus on a particle surface requires the microscope to move upward to a new position on the z axis. The distance between the two microscope positions is equal to the depth of a particle. The AIMS software sorts the three dimensions based on length and calculates the sphericity index as shown in Eq. (1): Sphericity = 3
d s .d l d2 L

(1)

where dL is the longest dimension, dI is the intermediate dimension, and ds is the shortest dimension. A sphericity value of one indicates that a particle has equal dimensions.
Angularity

The angularity is analyzed using the gradient method. This method quantifies the change in the gradient on a particle boundary. A higher gradient value indicates more angularity. The gradient method starts by calculating the inclination of gradient vectors on particle boundary points from the x-axis (horizontal axis in an image). The average change in the inclination of the gradient vectors is taken as an indication of angularity as follows: Angularity Index (Gradient Method) =

N 1 i =1 3

N 3

i i +3

(2)

where the subscript i denotes the i th point on the boundary of a particle, and N is the total number of points on the boundary. The average rather than the summation is considered in Eq. (2) so that the angularity calculation is not biased by particle size.

589

Texture

The wavelet method is used to quantify texture. The advantages of using this method over other methods such as the Fourier analysis are discussed by Mallat (1989). Mainly, the wavelet method has the advantage of decomposing an image into different levels. Consequently, each level is analyzed to quantify a certain texture scale. Wavelets have the advantage of capturing the sharp changes in texture in an image since its basis functions have variable durations that can fit these sharp changes. However, the

harmonic functions that constitute the basis for the Fourier analysis do not have limited duration, and they are not efficient in modeling these abrupt changes in texture on an image (Mallat 1989) As discussed by Chandan et al. (2004), the wavelet analysis gives the texture details in the horizontal, vertical, and diagonal directions in three separate images. The texture index is taken at a given decomposition level as the arithmetic mean of the squared values of the wavelet coefficients for all three directions. The texture index is expressed mathematically as follows: Texture Index =
1 3N

( D ( x, y ) )
i =1 j =1 i, j

(4)

where D is the wallet coefficient, N is the total number of coefficients, i takes a value 1,2 or 3, for the three directions of texture, and j is the wavelet coefficient index.
ANALYSIS AND RESULTS

All shape characteristics (angularity, texture and sphericity) were measured in each of the aggregate sizes present in the Micro-Deval sample before and after polishing. The detailed analysis results for each of the aggregate sources and sizes are available in 590

the directory labeled Aggregate Detailed Results in the enclosed CD. Each of the analysis files includes a summary sheet that shows the average and standard deviation of each of the measured characteristics. It also includes the percentage of aggregates that belong to each of the aggregate shape groups. The details on the development of these aggregate shape groups are provided by Al-Rousan et al. (2005). This summary report will focus on the analysis of the measured characteristics averaged for the sizes in an aggregate sample. These average values are shown in Table 1 and given in the file Aims Summary.xls, which is on the CD included with this report. Table 1: Average values of shape characteristics before (BMD) and after (AMD) MicroDeval abrasion.
Texture Aggregate AL-3 MB-2 MB-3 ME-1 ME-2 NB-1 NC-3 NC-4 NM-1 NM-3 NM-4 NY-5 ON-2 ON-3 ON-6 ON-7 ON-10 ON-13 OR-1 QC-1 BMD 124.39 189.63 73.75 239.87 236.12 248.21 270.42 226.18 253.27 125.04 228.50 122.17 294.75 276.63 167.96 229.28 356.36 214.10 202.12 309.86 AMD 130.57 139.33 60.69 209.07 190.35 128.96 126.95 196.25 317.34 100.72 152.88 64.59 294.41 313.34 119.95 142.85 186.72 88.38 214.55 333.77 Angularity BMD 3042.83 2398.93 1842.97 3603.77 3080.83 3088.22 3015.00 3996.31 3718.71 3480.43 2906.74 2784.90 3238.30 3287.57 3014.40 3069.39 3094.32 2625.89 3386.49 3548.26 AMD 2306.05 1767.58 1702.28 2317.60 2239.24 2022.75 2028.78 2464.27 2468.29 1948.41 2132.25 1797.21 2524.04 2179.97 1845.44 2038.82 2014.71 1634.47 2441.16 2106.64 Sphericity BMD 0.67 0.74 0.77 0.70 0.67 0.67 0.62 0.74 0.73 0.73 0.72 0.65 0.68 0.69 0.75 0.67 0.74 0.73 0.72 0.74 0.64 0.75 0.76 0.72 0.68 0.66 0.61 0.71 0.73 0.73 0.67 0.63 0.69 0.73 0.75 0.70 0.74 0.73 0.71 0.72 Percent Change Percent Change Percent Change in Angularity -24.21 -26.32 -7.63 -35.69 -27.32 -34.50 -32.71 -38.34 -33.63 -44.02 -26.64 -35.47 -22.06 -33.69 -38.78 -33.58 -34.89 -37.76 -27.91 -40.63 in Sphericity -3.78 0.67 -0.86 2.11 1.51 -0.59 -2.15 -3.69 1.16 0.08 -7.18 -3.16 1.62 5.89 0.66 4.25 0.66 1.32 -0.96 -2.34 4.97 -26.53 -17.71 -12.84 -19.38 -48.04 -53.06 -13.23 25.29 -19.45 -33.09 -47.13 -0.11 13.27 -28.59 -37.70 -47.60 -58.72 6.15 7.72

AMD in Texture

591

Figure 1 shows a comparison between the average angularity values before and after abrasion in the Micro-Deval. It can be seen that all samples lost some angularity. This can be seen clearly in Figure 2 in which a negative value indicates a loss of angularity due to abrasion. The percentage change in angularity is defined as the

angularity after Micro-Deval minus the angularity before Micro-Deval divided by the angularity before Micro-Deval. The loss of angularity ranged between 7.5% to about 44%.

5000 4500 4000 Angularity Index After MD 3500 3000 2500 2000 1500 1000 500 0 0 1000 2000 3000 4000 5000 Angularity Index Before MD Equality Line

Figure 1: Comparison between angularity before and after Micro-Deval abrasion.

592

0.00 -5.00 Percent Change in Angularity, % -10.00 -15.00 -20.00 -25.00 AL-3 MB-2 -30.00 -35.00 ME-1 -40.00 -45.00 -50.00 Aggregate Source ON-2 ME-2 NC-3 NB-1 NM-4 OR-1 ON-3 ON-7 ON-10 ON-13 QC-1 MB-3

NM-1

NY-5

NC-4 NM-3

ON-6

Figure 2: Percent change in angularity due to Micro-Deval abrasion.

The texture results are shown in Figures 3 and 4. It is interesting to note that not all aggregates lost texture due to abrasion in the Micro-Deval. This phenomenon has actually been reported in previous studies (Mahmoud 2005). The increase in texture has been noticed in porous aggregates or siliceous river gravel in which more textured surfaces are exposed due to abrasion.

593

400 Equality Line 350 300 Texture Index After MD 250 200 150 100 50 0 0 50 100 150 200 250 300 350 400 Texture Index Before MD

Figure 3: Comparison between texture before and after Micro-Deval abrasion.

594

30.00 20.00 Percent Change in Texture, % 10.00 0.00 -10.00 -20.00 -30.00 -40.00 -50.00 -60.00 -70.00 NB-1 NC-3 ME-1 MB-3 ME-2 MB-2 AL-3

NM-1 ON-3 OR-1 QC-1

ON-2

NC-4 NM-3 ON-6 NM-4 ON-7 NY-5 ON-10 ON-13

Aggregate Source

Figure 4: Percent change in texture due to Micro-Deval abrasion.

The sphericity results are reported in Figures 5 and 6. It can be seen in these figures that the Micro-Deval abrasion has almost no effect on aggregate shape. The maximum change in sphericity was 8%. This can be attributed to the fact that abrasion causes uniform change in aggregate dimensions which leaves the sphericity index almost unchanged.

595

1 0.9 0.8 Sphericity Index After MD 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 0.2 0.4 0.6 0.8 1 Sphericity Index Before MD Equality Line

Figure 5: Comparison between sphericity before and after Micro-Deval abrasion.

596

8.00 6.00 Percent Change in Sphericity, % 4.00 2.00 MB-2 0.00 -2.00 -4.00 -6.00 -8.00 NM-4 Aggregate Source MB-3 NB-1 NC-3 AL-3 NC-4 NY-5 OR-1 QC-1 ME-1 ME-2 ON-2 ON-6 ON-3 ON-7

NM-1 NM-3

ON-13 ON-10

Figure 6: Percent change in sphericity due to Micro-Deval abrasion.

In the Micro-Deval test, aggregates are subjected to both abrasion and breakage, and both of these mechanisms are associated with weight loss. Visual inspection of aggregates after Micro-Deval testing indicated that some of the aggregates were only abraded, while others experienced breakage with minimal change in their surface angularity. Mahmoud (2005) developed a procedure to distinguish between aggregate breakage and abrasion. This procedure consists of three steps: (1) measure aggregate initial angularity, (2) test the aggregate in the Micro-Deval, and (3) measure its angularity and weight loss after the Micro-Deval.

597

The percent change in angularity is plotted against Micro-Deval weight loss (aggregate passing sieve #16) in Figure 7 to distinguish between abrasion and breakage. Aggregates with high weight loss but low angularity loss were those that experienced high breakage and low abrasion. Aggregates that had high angularity loss and high weight loss were the ones that encountered both high abrasion and high breakage. On the other hand, low values of weight loss and angularity loss were associated with low abrasion and breakage. Finally, aggregates with high angularity loss but low weight loss were the ones that had high abrasion and low breakage.

60

High Abrasion Low Breakage

High Abrasion High Breakage

50 Percent Loss in Angularity, %

40

30

20

10

Low Abrasion Low Breakage

0.0 5.0 10.0 15.0

Low Abrasion High Breakage

20.0 25.0 30.0 35.0 40.0 Percent Weight Loss, %

Figure 7: Percent weight loss in Micro-Deval versus percent change in angularity.

598

SUMMARY

The Aggregate Imaging System (AIMS) was used to measure the shape characteristics (angularity, texture and form) of aggregate samples before and after abrasion in the Micro-Deval. It is suggested the ICAR researchers correlate the following average parameters to the reported performance of these aggregate samples:

Initial texture and angularity. Percent change in angularity and texture. The resistance to abrasion and breakage as analyzed in Figure 7 of this report.

The Aggregate Detailed Results directory includes the details of the distribution of shape characteristics in each of the aggregate samples. It is highly recommended that the researchers explore the relationships between the distributions of shape characteristics with the reported performance. It is possible that performance is more influenced by the distribution of shape characteristics rather than the average values of these characteristics.
REFERENCES
Al-Rousan, T., Masad, E., Myers, L., and Spiegelman, C. (2005). A New Methodology for Shape Classification of Aggregates, In Transportation Research Record 1913, Journal of the Transportation Research Board, pp. 11-23.

Chandan, C., Sivakumar, K., Fletcher, T., and Masad, E. (2004). Geometry Analysis Of Aggregate Particles Using Imaging Techniques. Journal of Computing in Civil Engineering, ASCE, Vol. 18, No. 1, pp. 75-82. Mahmoud, E. (2005). Development of Experimental Methods for the Evaluation of Aggregate Resistance to Polishing, Abrasion, and Breakage, M.Sc. Thesis, Department of Civil Engineering, Texas A&M University, College Station, TX.

599

Mallat, S.G. A Theory for Multiresolution Signal Decomposition: The Wavelet Representation. (1989). IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 11, pp. 674-693 Masad, E., Al-Rousan, T., Button, J., Little, D., and Tutumluer, E. (2005). Test Methods for Characterizing Aggregate Shape, Texture, and Angularity, Final Report of Project 430A, National Cooperative Highway Research Program, Transportation Research Board, National Research Council, Washington, D.C.

600

References

American Association of State Highway and Transportation Officials, "Standard Test Method for Soundness of Aggregate by Use of Sodium Sulfate or Magnesium Sulfate," T 104-99, 1999. American Association of State Highway and Transportation Officials, "Standard Test Method for Resistance of Coarse Aggregate to Degradation by Abrasion in the MicroDeval Apparatus," TP58-00, 1999. American Association of State Highway and Transportation Officials, "Standard Method of Test for Soundness of Aggregates by Freezing and Thawing," T 103-91, 2000. Ahlrich, R. C., "Influence of Aggregate Properties on Performance of Heavy-duty Hotmix Asphalt Pavements," Transportation Research Record, No. 1547, National Research Council, Washington, DC, USA, 1996, p. 7. Amirkhanian, S., D. Kaczmarek, and J. Burati, "Effects of Los Angeles Abrasion Test Values on the Strengths of Laboratory-Prepared Marshall Specimens," Transportation Research Record, No. 1301, 1991, pp. 77-86. Amirkhanian, S. N., E. L. Dukatz, Jr., D. Kaczmarek, and D. Brownlow, "Effects of Igneous Aggregate Sources with Various Los Angeles Abrasion Test Values on the Strengths of Concrete Mixtures," Cement, Concrete and Aggregates, Vol. 14, No. 2, 1992, p. 86. Benson, P. E., and W. H. Ames, "The Precision of Selected Aggregate Test Methods," Report No. CA-DOT-TL-1153-2-75-05, California Department of Transportation, 1975. Bloem, D. L., "Soundness and Deleterious Substances," Significance of Tests and Properties of Concrete and Concrete Making Materials, ASTM STP 169A, No. 169, American Society for Testing and Materials, 1966, pp. 497-512. Boucher, D. L., and E. T. Selig, "Application of Petrographic Analysis to Ballast Performance Evaluation," Transportation Research Record, No. 1131, 1987, pp. 15-25. British Standards Institution, "Methods for Determination of aggregate crushing value (ACV)," BS 812: Part 110, 1990. British Standards Institution, "Methods for Determination of aggregate impact value (AIV)," BS 812: Part 112, 1990. 601

Cooley, L. A., and R. S. James, "Micro-Deval Testing of Aggregates in the Southeast," Transportation Research Record, No. 1837, National Research Council, 2003, p. 73. Cooper, A. J., J. S. Villard, and J. D. Ferguson, "Predicting Aggregate Performance of the Iguneous and Metamorphic Rocks in Northern Ontario," International Center for Aggregates Research 11th Annual Symposium, Austin, TX, 2003. Crenshaw, M., Personal Interview, Austin, TX, April 29, 2004. Depuy, G. W., "Petrographic Investigations of Concrete and Concrete Cggregates at the Bureau of Reclamation," ASTM STP 1061, Publ by ASTM, Philadelphia, PA, USA, 1990, pp. 32-46. Desolminihae, H.E., W.R. Hudson, and E. Ricci, "Serviceability Ratings of Texas Highway System for Pavement Management," Research Report 400-1F, Center for Transportation Research, The University of Texas at Austin, August 1986. Dhir, R. K., D. M. Ramsay, and N. Balfour, "Study of the Aggregate Impact and Crushing Value Tests," The Journal of the Institution of Highway Engineers, Vol. 18, No. 11, 1971, p. 27. Ekse, M., and H. C. Morris, "A Test for Production of Plastic Fines in the Process of Degradation of Mineral Aggregates," Symposium on Road and Paving Materials, ASTM STP 277, Philadelpha, 1959, p. 122. Embacher, R. A., and M. B. Snyder, "Refinement and Validation of the Hydraulic Fracture Test," Transportation Research Record, No. 1837, National Research Council, 2003, p. 80. Esa, S., H. Hozayen, and A. Garif, "Estimating Aggregate Water Absorption from the Los Angeles Abrasion test Results," Proceedings of the 1993 Annual Conference of the Transportation Association of Canada, Vol. 1, Ontario, Canada, 1993. Fistric, M., I. Tomasic, and M. Vrkljan, "Influence of Petrographic Characteristics of Silicate Rocks on the Quality of Aggregates," Rudarsko Geolosko Naftni Zbornik, No. 14, University of Zagreb, 2002, p. 11. Forster, S. W., "Soundness, Deleterious Substances, and Coatings," Significance of Tests and Properties of Concrete and Concrete Making Materials, ASTM Special Publication, STP 196C, American Society for Testing and Materials, 1994, p. 411. Gandhi, P. M., and R. L. Lytton, "Evaluation of Aggregates for Acceptance in Asphalt Paving Mixtures," Proceedings of the Association of Asphalt Paving Technologists, Vol. 53, Scottsdale, AZ, 1984, pp. 525-547. 602

Garrity, L. V., and H. F. Kriege, "Studies of the Accelerated Soundness Tests," Proceedings, Highway Research Record, Vol. 15, 1935, pp. 237-260. Gilbert, K., Micro-Deval Study, Unpublished Internal Report, Colorado Department of Transportation, 2003. Goldsworthy, S., Personal Interview, Austin, TX, August 10, 2004. Goonewardane, K., "Behavior of Aggregate in the Washington Degradation Test," Journal of Testing and Evaluation, Vol. 5, No. 1, 1977, p. 25. Hanna, A., K. Folliard, and K. Smith, "Aggregate Tests for Portland Cement Concrete Pavements: Review and Recommendations," Research Results Digest, No. 281, NCHRP, 2003. Harman, J. W., P. D. Cady, and N. B. Bolling, "Slow-Cooling Test for Frost Sysceptibility of Pennsylvania Aggregates," Highway Research Record, No. 328, 1970, pp. 26-37. Hobson, K., Z. Musharraf, and R. A. Tarefder, "Micro-Deval Test for Evaluating Properties of Roadway Aggregates," International Journal of Pavements, Vol. 2, No. 1-2, 2003, pp. 8-19. Hornych, P., "COST 337 - Unbound Granular Materials for Roads," Working Group 2A Review of Tests, Test Procedures, and Methodolgy of Studies, Laboratoire Central des Ponts et Chaussees, 1998. Hudec, P. P., "Aggregate Tests - Their Relationship and Significance," Durability of Building Materials, Vol. 1, No. 3, 1983, p. 275. Hunt, E. A., "Micro-Deval Coarse Aggregate Test Evaluation," Report No. OR-RD-0113, Oregon Department of Transportation, 2001. Hveem, F. N., and T. W. Smith, "A Durability Test for Aggregates," Highway Research Record, No. 62, 1964, pp. 119-136. Issa, M. A., and M. A. Issa, "Evaluation of Washington Hydraulic Fracture Test (SHRP) for D-Cracking Aggregtes," Report No. ITRC FR 95/96-5, Illinois Transportation Research Center, 1999. Jayawickrama, P. W., W. F. Phillips, and M. S. Hossain, "Comparative Analysis of the Micro-Deval and Magnesium Sulfate Soundness Tests," Report No. 1771-S, Texas Tech University, Center for Multidisciplinary Research in Transportation, 2001.

603

Jumper, E. A., J. D. Herbert, and C. W. Beardsley, "Rattler Losses Correlated with Compression Strength of Concrete," Journal, American Concrete Institute, 1956, p. 563. Kandhal, P., and F. Parker, "Aggregate Tests Related to Asphalt Concrete Performance in Pavement," Report No. 405, NCHRP, 1998. Kohler, G., and J. Nagel, "Comparative Investigations of International Test Methods for Small-Sized Coarse Aggregates," Highway Research Board Record, No. 412, 1972, p. 51. Korhonen, C., and B. Charest, "Assessing Cryogenic Testing of Aggregate for Concrete Pavements," Report No. 95-4, US Army Corps of Engineers, Hanover, NH, USA, 1995. Landgren, R., "Unit Weight, Specific Gravity, Absorption, and Surface Moisture," Significance of Tests and Properties of Concrete and Concrete Making Materials, ASTM Special Publication, STP 196C, American Society for Testing and Materials, 1994, p. 421. Laplante, P., P. C. Aitcin, and D. Vezina, "Abrasion Resistance of Concrete," Journal of Materials in Civil Engineering, Vol. 3, No. 1, 1991, p. 19. Larson, L. J., R. P. Mathiowetz, and J. H. Smith, "Modification of the Standard Los Angeles Abrasion Test," Highway Research Record, No. 353, 1971, pp. 15-24. Larson, T., P. Cady, M. Franzen, and J. Reed, "Special Report No. 80," Highway Research Board, 1964. Latham, J. P., and C. Rogers, "Canadian experience with the micro-Deval test for aggregates," Geological Society Engineering Geology Special Publication, No. 13, 1998, pp. 139-47. Latham, J. P., A. R. Woodside, and W. D. H. Woodward, "Assessing the Wear Characteristics of Aggregate Exposed at the Road Surface," Geological Society Engineering Geology Special Publication, No. 3, 1998, pp. 149-157. Litts, H. W., "Evaluation Report: Magnesium Sulfate Soundness of Granular Materials: Temperature and Sample Washing," New York State Department of Transportation, 1981. Litvan, "Frost Action in Cement in the Presence of De-icers," Cement and Concrete Research, Vol. 6, 1976, pp. 351-356. Loughlin, G. F., "Usefulness of Petrology in the Selection of Limestone," Rock Products, Vol. 31, 1928, p. 50. 604

Marks, V. J., and W. Dubberke, "Durability of Concrete and the Iowa Pore Index Tests," Transportation Research Record, No. 853, 1982, p. 25. Mccown, V., "The Significance of Sodium Sulfate and Freezing and Thawing Tests of Mineral Aggregates," Proceedings, Vol. 11, No. 1, Highway Research Board, 1932, pp. 312-337. Meininger, R., "Validity of the Sulfate Soundness Test," Rock Products (www.rockproducts.com), Accessed January 8, 2004; Posted August 1, 2002. Meininger, R., "Micro-Deval vs. L. A. Abrasion," Rock Products, Vol. 107, No. 4, Primedia Intertec Publishing Corp., 2004, p. 33. Meininger, R. C., "Degradation Resistance, Strength, and Related Properties," Significance of Tests and Properties of Concrete and Concrete Making Materials, ASTM Special Publication, STP 169C, American Society for Testing and Materials, 1994, p. 388. Melville, P. L., "Weathering Studies of Some Aggregates," Proceedings, Vol. 28, Highway Research Board, 1948, p. 238. Mielenz, R. C., "Petrographic Evaluation of Concrete Aggregates," Significance of Tests and Properties of Concrete and Concrete Making Materials, ASTM Special Publication, STP 169C, American Society for Testing and Materials, 1994, p. 341. Miles, D. K., "Accelerated Soundness Test for Aggregates," Report No. 500-917, Utah State Department of Highways - Materials and Testing Division, 1972. Minor, C. E., "Degradation of Mineral Aggregate," Symposium on Road and Paving Materials, ASTM STP 277, Philadelphia, 1959, p. 109. Mirza, W., "Evaluation Of Jeddah Aggregates For Concrete Construction," Journal of Engineering and Applied Sciences, Vol. 3, 1986, pp. 67-72. Moavenzdeh, F., and W. H. Goetz, "Aggregate Degradation in Bituminous Mixtures," Highway Research Board Record, No. 24, 1963, pp. 106-136. Nokken, M. R., R. D. Hooton, and C. A. Rogers, "Measured Internal Temperatures in Concrete Exposed to Outdoor Cyclic Freezing," Cement, Concrete and Aggregates, Vol. 26, No. 1, American Society for Testing and Materials, 2004, p. 26. Ontario Ministry of Transportation, "Method of Test for Resistance of Unconfined Coarse Aggregate to Freezing-and-Thawing," LS-614, 1989.

605

Papaleontiou, C. G., A. H. Meyer, and D. W. Fowler, "Evaluation of the 4-Cycle Magnesium Sulfate Soundness Test," Report No. CTR 3-9-85-438-IF, Center for Transportation Research, The University of Texas at Austin, 1987. Paul, I., "Magnesium Sulphate Accelerated Soundness Test on Concrete on Aggregates," Proceedings of the Twelfth Annual Meeting of the Highway Research Board, Vol. 12, 1932, p. 319. Pigeon, M., J. Prevost, and J.-M. Simard, "Freeze-Thaw Durability vs. Freezing Rate," Journal of The American Concrete Institute, Vol. 82, No. 5, 1985, p. 684. Powers, T. C., "Basic Considerations Pertaining to Freezing-and-Thawing Tests," Proceedings, Vol. 55, American Society for Testing and Materials, 1955, pp. 1121-1155. Ramsay, D. M., "Factors Influencing Aggregate Impact Value in Rock Aggregate," The Quarry Managers' Journal., 1965, pp. 129-134. Ramsay, D. M., R. K. Dhir, and J. M. Spence, "The Practical and Theoretical Merits of the Aggregate Impact Value in the Study of Crushed Rock Aggregate," Proceedings of a Conference on Rock Engineering, The University of Newcastle upon Type, England, 1977. Rhoades, R., and R. C. Mielenz, "Petrographic and Mineralogic Characteristics of Aggregates," Mineral Aggregates, ASTM STP 83, American Society for Testing and Materials, 1948, pp. 20-48. Richard, J. A., and J. R. Scarlett, "Airport Engineering ATR-024: A Review and Evaluation of the Micro-Deval Test," Public Works and Government Services of Canada, Report on Project 914222, 1997. Rogers, C., "Search for Skid Resistance Aggregates in Eastern Ontario: Interim Report," Report No. EM-036, Ontario Ministry of Transportation, 1980. Rogers, C. A., S. A. Senior, and D. Boothe, "Development of an Unconfined FreezeThaw Test for Coarse Aggregates," Technical Report EM-97, Engineering Materials Office, Ontario Ministry of Transportation, 1989. Rogers, C. A., "Petrographic Examination of Aggregate and Concrete in Ontario," ASTM STP 1061, Publ by ASTM, Philadelphia, PA, USA, 1990, p. 5. Rogers, C. A., M. L. Bailey, and B. Price, "Micro-Deval Test for Evaluating the Quality of Fine Aggregate for Concrete and Asphalt," Transportation Research Record, No. 1301, 1991, pp. 68-76.

606

Rogers, C. A., and S. A. Senior, "Recent Developments in Physical Testing of Aggregate to Ensure Durable Concrete," Advances in Cement and Concrete, American Society for Testing and Materials, 1994, pp. 338-357. Rogers, C. A., S. I. Szoke, and R. G. Gorman, "Granular Base Failures in Low-Volume Roads in Ontario, Canada," 6th International Conference on Low-Volume Roads, Minneapolis, MN, 1995, pp. 23-32. Rogers, C. A., B. C. Lane, and S. A. Senior, "The Micro-Deval Abrasion Test for Coarse and Fine Aggregate in Asphalt Pavement," International Center for Aggregates Research 11th Annual Symposium, Austin, TX, 2003. Rogers, C. A., Personal Interview, Austin, TX, July 27, 2004. Senior, S. A., and C. A. Rogers, "Laboratory Tests for Predicting Coarse Aggregate Performance in Ontario," Transportation Research Record, No. 1301, 1991, pp. 97-106. Shakoor, A., T. West, and C. Scholer, "Physical Characteristics of Some Indiana Argillaceous Carbonates Regarding Their Freeze-Thaw Resistance in Concrete," Bulletin of the Association of Engineering Geologists, Vol. 19, No. 4, 1982, pp. 371-384. Shakoor, A., and C. L. Brown, "Development of a Quantitative Relationship Between Unconfined Compressive Strength and Los Angeles Abrasion Loss for Carbonate Rocks," Bulletin of the International Association of Engineering Geology, No. 53, 1996, p. 103. Sheftick, W., "Na2SO4 Soundness Test Evaluation," Cement, Concrete and Aggregates, Vol. 11, No. 1, ASTM, 1989, p. 73. Sheftick, W. M., Personal Interview, Austin, Texas, April 12, 2005. Shergold, F. A., "A Review of Available Information on the Significance of Roadstone Tests," Report No. 10, Road Research Laboratory, Department of Scientific and Industrial Research, United Kingdom, 1948. Smith, F. L., "Effect of Aggregate Quality on Resistance of Concrete to Abrasion," Cement and Concrete, ASTM STP 205, American Society for Testing and Materials, 1958, p. 91. Spellman, D. L., J. H. Woodstrom, and S. N. Bailey, "Concrete Aggregate Durability Tests," Report No. CA-DOT-TL-3476-2-75-44, California Department of Transportation Laboratory, 1975, p. 45. Standards Association of Australia, "Wet/Dry Strength Variation," Methods for Sampling and Testing Aggregates, Australian Standard, AS 1141.22, 1980. 607

Stark, D., "Characteristics and Utilization of Coarse Aggregates Associated with DCracking," Living with Marginal Aggregates, ASTM STP 597, American Society for Testing and Materials, 1976, p. 45. Swenson, E. G., and V. Chaly, "Basis for Classifying Deleterious Characteristics of Concrete Aggregate Materials," Proceedings, Vol. 52, American Concrete Institute, 1956, pp. 987-1002. U.S. Army Corps of Engineers, "Investigation of a Proposed Aggregate Degradation Test Method," Miscellaneous Paper No. 6-218, U.S. Army Corps of Engineers, Vicksburg, MS, 1957. Verbeck, G., and R. Landgren, "Influence of Physical Characteristics of Aggregates on Frost Resistance of Concrete," Proceedings, Vol. 60, American Society for Testing and Materials, 1960, pp. 1063-1079. Volger, R., and G. H. Grove, "Freeze-Thaw Testing of Coarse Aggregate in Concrete: Procedures Used by Michigan Department of Transportation and Other Agencies," Cement, Concrete and Aggregates, Vol. 11, No. 1, American Society for Testing and Materials, 1989, pp. 57-66. Walker, S., and C. E. Proudley, "Studies of Sodium and Magnesium Sulfate Soundness Tests," American Society for Testing and Materials, Vol. 36, 1936, p. 327. Winslow, D., "Pore System of Coarse Aggregates," Significance of Tests and Properties of Concrete and Concrete Making Materials, ASTM Special Publication, STP 196C, American Society for Testing and Materials, 1994, p. 429. Woolf, D. O., "Relation Between Sodium Sulfate Soundness Tests and Absorption of Sedimentary Rock," Public Roads, 1927, p. 225. Woolf, D. O., "The Relation Between Los Angeles Abrasion Test Results and the Service Records of Coarse Aggregates," Proceedings of the Seventeenth Annual Meeting of the Highway Research Board, Vol. 17, 1937, pp. 350-359. Woolf, D. O., "Improvement in the Uniformity of the Accelerated Soundness Test of Coarse Aggregate," Bulletin No. 213, American Society for Testing and Materials, 1953, p. 42. Wu, Y., F. Parker, and P. S. Kandhal, "Aggregate toughness/abrasion resistance and durability/soundness tests related to asphalt concrete performance in pavements," Transportation Research Record, No. 1638, National Research Council, Washington, DC, USA, 1998, p. 85. 608

Wu, Y., J. F. Parker, and P. S. Kandhal, "Evaluation of tests for Toughness/Abrasion Resistance and Durability/Soundness of Coarse Aggregate," Cement, Concrete, and Aggregates, Vol. 21, ASTM, 1999, pp. 12-22. Wuerpel, C. E., "Factors Affecting the Testing of Concrete Aggregate Durability," Proceedings, Vol. 38, No. 1, American Society for Testing and Materials, 1938, p. 327. Wuerpel, C. F., "Modified Procedure for Testing Aggregate Soundness by Use of Magnesium Sulfate," Proceedings, Vol. 39, American Society for Testing and Materials, 1939, p. 882.

609