The Development of a Standard Test Method for ...

Development of a Standard Test Method for Determining the Bitumen Bond Strength of Emulsions A South African Perspective

by Andries Hendrik Greyling

Thesis presented in fulfilment of the requirements for the degree of Master of Science in Engineering in the Faculty of Civil Engineering at Stellenbosch University

Supervisor: Professor Kim Jenkins

December 2012

Development of a Standard Test Method for Determining Bitumen Bond Strength of Emulsions A South African Perspective

Declaration

By submitting this thesis electronically, I declare that the entirety of the work contained therein is my own, original work, that I am the owner of the copyright thereof (unless to the extent explicitly otherwise stated) and that I have not previously in its entirety or in part submitted it for obtaining any qualification.

Andries Hendrik Greyling

Date: 20 November 2012

Copyright © 2012 Stellenbosch University

All rights reserved

ii AH GREYLING

December 2012


Abstract Chip and spray surfacing seals has been widely used in South Africa as the preferred surface treatment for rural roads. The design of these seals has also experienced renewed interest and continuous development in several other countries over the past two decades. In South Africa seals are continually used as increasing attention is given to the periodic maintenance of existing surfaced roads. There is also a significant increase in the use of surfacing seals in North America as the need to develop more energy and resource efficient surfacing options becomes a priority.

Despite this growing use of surface seals, the seal design and especially the selection of binder type and grade does not always follow scientific processes. Seals are often designed based on client preference, previous experience, material availability and industry trends. With an ageing road network and limited funding to ensure timeous maintenance, the focus in South Africa is moving towards more stringent and scientific design processes.

The USA and Europe, forced by increasing traffic volumes and heightened performance demands, are moving towards performance- based specification to account for deficiencies in their current surface seal design methods. One of the major failure mechanisms of surface seals is ravelling which takes place when the binder and the aggregate bond becomes weak and are broken by the forces generated by traffic. This leads to loose aggregate on the road which in turns leads to bare surface patches and broken windscreens. To prevent and address this as part of the development of performance-based specifications, the need for a simple and inexpensive technique for evaluating bitumen and bitumen emulsion bond strength development over time, as well as binder-aggregate compatibility, was identified.

Although various tests exist for investigating adhesion between bituminous emulsions and aggregate chips most of the tests does not deliver the level of information required by the performance-based specifications. The Bitumen Bond Strength (BBS) test method was therefore developed with the aim to address some of the limitations encountered in evaluating bond strength between binders and aggregates. The BBS test (AASHTO TP-91, 2011) was developed by the University of Wisconsin – Madison (UWM) in partnership with the University of Ancona – Italy (UAI) and the University of Stellenbosch – South Africa (US) specifically for evaluating bond strength between aggregates and hot applied binders and emulsions, respectively.

US became involved in BBS test efforts in 2008 to assist in the development and practical evaluation of the BBS test method. UAI contributed significantly to the development of the test apparatus through their work in conjunction with UWM. Due to time and resources available, the involvement of the US was limited to various discussion sessions, a study tour, the evaluation of the BBS test, and conducting a series of control tests. iii AH GREYLING

December 2012


By using the Bitumen Bond Strength test it was possible to evaluate the practicality and repeatability of the developed procedure and the results measured was successfully used to evaluate the bond strength development of modified and unmodified bitumen emulsions on tillite and granite aggregates. It was also possible to correlate the results achieved at the University of Stellenbosch with results from the University of Wisconsin-Madison due to the fact that testing took place at both institutions.

The development of the test and the inter-laboratory test results in essence reinforced the hypothesis that the BBS test protocol can be used to effectively evaluate bond strength of different emulsion types and aggregate types. Except for the loading rate which is a known critical influence, the emulsion type and curing intervals are both identified as the most significant other factors contributing to bond strength development.

Aggregate type is also identified as a significant factor that will influence the bond strength development. Interactions between emulsion type and curing interval are identified as the most significant interaction. A lot of further validation test on the BBS test method is still required for the test to be integrated into a performance-based specification system for surface seals.

iv AH GREYLING

December 2012


Opsomming “Chip and Spray” oppervlak seëls is die verkose seël tipe vir gebruik op Suid Afrikaanse plattelandse paaie. Die ontwerp van hierdie tipe seëls is ook tans besig om hernude aandag te trek in heelwat lande. In Suid Afrika word die tipe seël al meer gebruik soos die behoefte na herseël projeke op die verouderde pad netwerk groei . Daar is ook heelwat groei in die gebruik van “Chip en Spray” seëls in Noord Amerika aangesien daar n behoefte is om n meer energie en materiaal effektiewe seëls te gebruik.

Ten spyte van die groeiende gebruik van hierdie seëls het daar min wetenskaplike ontwikkeling plaasgevind in the ontwerp daarvan. Seëls word meestal ontwerp gebaseer op klient voorkeur , ondervinding, materiaal beskikbaarheid en ook industrie norms. ‘n Verouderde pad netwerk dwing die Suid Afrikaanse industrie om meer deeglike en wetenskaplike ontwerp prosedures te volg.

Die VSA en Europa word deur vinnig groeiende verkeersvolumes en verhoogte kwaliteits behoeftes gedwing om prestasie gebaseerde spesifikasies te ontwikkel. Een van die hoof defekte op seëls is klipverlies wat plaasvind as die verbinding tussen die bitumen en aggregaat verswak en gebreek word deur die kragte wat deur verkeer oorgedra word. Dit lei tot los klip op die pad wat weer tot skade aan voertuie lei. Om dit te voorkom en aan te spreek het die behoefte laat ontwikkel vir n maklike en goedkoop tegniek om te meet hoe sterk die verbinding is wat ontwikkel tussen verskillende bitumen, bitumen emulsies en aggregaat monsters.

Alhoewel daar alreeds toetse bestaan wat kan toets hoe sterk die verbinding is , gee die meeste van die toetse nie die informasie wat benodig word deur die voorgestelde prestasie gedrewe spesfikasies nie. Die “Bitumen Bond Strength” (BBS) toets metode is daarom ontwikkel om die tekortkoming in die toets van die bitumen en aggregaat verdindinge aan te spreek.

Die BBS toets (AASHTO TP-91, 2011) is ontwikkel by die Universiteit van Wisconsin – Madison (UWM) in vennootskap met die Universiteit van Ancona – Italy (UAI) en die Universiteit van Stellenbosch (US) , spesifiek om die die sterkte van die verbinding wat vorm tussen aggregaat en bitumen te meet.

Die US het in 2008 betrokke geraak by die BBS toets studie om hulp te verleen met die ontwikkeling en praktiese evalueering van die BBS toets metode. As gevolg van tyd en personeel tekorte is die betrokkendheid by die US v AH GREYLING

December 2012


beperk tot verskeie besprekings sessies, n studie toer, die evalueering van die BBS toets sowel as die voltooing van n uitgebreide stel toetse.

Deur die voltooing van eksperimente met die BBS toets was dit moontlik om die uitvoerbaarheid en herhaalbaarheid van die ontwikkelde prosedure te toets. Die resultate van die toetse is suksesvol gebruik om die ontwikkeling van die verbinding sterkte tussen gemodifiseerde en ongemodifiseerde bitumen emulsies en tilliet en graniet te definïeer en te evalueer. Dit was ook moontlik om die resultate van die Universiteit Stellenbosch en die Universiteit of Wisconsin-Madison suksesvol met mekaar te vergelyk aangesien toetse by beide die instansies voltooi is.

Die ontwikkeling van die toets en die inter laboratorium toets resultate het dit moontlik gemaak om die hipotese te bevestig dat die BBS toets prosedure effektief gebruik kan word om die bitumen en aggregaat verdindinge te toets en te evalueer. Behalwe vir die tempo van die lading waarteen die aftrek sterkte getoets word , is die emulsie tipe en die nabehandeling tydperk beide geidentifiseer as die mees beduidende invloede wat bydrae tot die ontwikkeling van die verbinding sterkte.

Die aggregaat tipe is ook geidentifiseer as n belangrike faktor wat die verbinding sterkte ontwikkeling sal beïnvloed. Die interaksie tussen die emulsie tipe en nabehandeling tydperk was geïdentifiseer as die mees beduidende interaksie. Daar sal wel nog heelwat eksperimente voltooi moet word met die BBS toets prosedure voordat dit volkome geïntegreer kan word as deel van n prestasie gebaseerde spesifikasie stelsel vir die ontwerp van seëls.

vi AH GREYLING

December 2012


Acknowledgements I gratefully acknowledge the following:



Professor Kim Jenkins, SANRAL Chair, University of Stellenbosch, for his mentorship, guidance and patience.



Mr Bryce Constable, Former Stellenbosch University Student & Friend, for his diligent completion of the endless tests.



Professor Hussain Bahia and Mr Timothy Miller, University of Wisconsin – Madison – USA. Their ground-breaking work made this study possible. I would like to thank them for giving me an opportunity to have a small part in it.



BVi Consulting Engineers, My Employer and Colleagues, for their financial support and the time to make this study possible.



Miss Hannalie Wiese for her patience, support and understanding



Mr Dries en Me. Lulu Greyling, My Parents, for their moral support and guidance.



To each and every one who was a part of this and who assisted in any way over the last four years. Thank you!

vii AH GREYLING

December 2012


TABLE OF CONTENT

1

2

3

INTRODUCTION ............................................................................................................................1 1.1

BACKGROUND ............................................................................................................... 1

1.2

PROJECT DEFINITION AND OBJECTIVES ................................................................ 2

1.2.1

PROJECT DEFINITION ................................................................................................... 2

1.2.2

RESEARCH OBJECTIVES .............................................................................................. 2

1.3

STUDY LOCATION ........................................................................................................ 3

1.4

METHODOLOGY ............................................................................................................ 4

1.5

THESIS OUTLINE ........................................................................................................... 4

AN INTRODUCTION TO ROADS AND ROAD SURFACING SEALS ...................................6 2.1

A BRIEF HISTORY OF ROADS ..................................................................................... 6

2.2

HISTORY OF BITUMEN SURFACING SEALS ......................................................... 7

2.2.1

REST OF THE WORLD.................................................................................................... 7

2.2.2

SOUTH AFRICA ............................................................................................................... 9

ROAD SURFACING SEALS ........................................................................................................11 3.1

INTRODUCTION ........................................................................................................... 11

3.2

SURFACING SEALS FUNCTION ................................................................................ 11

3.3

SURFACING SEAL FAILURE MODES AND MECHANISMS ................................. 12

3.3.1

BACKGROUND.............................................................................................................. 12

3.3.2

MODES OF DISTRESS .................................................................................................. 12

3.3.3

MODES OF SURFACE DISTRESS ............................................................................... 14

3.4

SURFACING SEAL DESIGN ........................................................................................ 17

3.4.1

INTRODUCTION............................................................................................................ 17

3.4.2

SOUTH AFRICAN METHOD, TRH3 ............................................................................ 18

3.4.3

REST OF THE WORLD.................................................................................................. 20

3.4.4

ROAD NOTE 39-UK ....................................................................................................... 23

3.4.5

AUSTROADS SPRAYED SEAL DESIGN METHOD .................................................. 24

3.5

GENERAL ...................................................................................................................... 25

3.6

FACTORS INFLUENCING THE DESIGN & PERFORMANCE OF SEALS ............. 25

3.6.1

INTRODUCTION............................................................................................................ 25 viii

AH GREYLING

December 2012


4

5

6

7

3.6.2

SUBSTRATE ................................................................................................................... 26

3.6.3

TRAFFIC ......................................................................................................................... 27

3.6.4

ROAD GEOMETRY ....................................................................................................... 28

3.6.5

CLIMATE AND ENVIRONMENTAL INFLUENCES ................................................. 28

3.6.6

AGGREGATES ............................................................................................................... 29

3.6.7

BITUMEN BINDERS ..................................................................................................... 30

BITUMEN.......................................................................................................................................36 4.1

INTRODUCTION ........................................................................................................... 36

4.2

MANUFACTURE OF BITUMEN ................................................................................. 36

4.3

COMPOSITION OF BITUMEN .................................................................................... 37

4.4

TYPES AND GRADES OF BITUMEN ......................................................................... 37

4.4.1

PENETRATION GRADE BITUMEN’S ......................................................................... 38

4.4.2

CUTBACK BITUMEN ................................................................................................... 38

4.4.3

BLOWN BITUMEN ........................................................................................................ 39

4.4.4

BITUMEN EMULSIONS ................................................................................................ 39

4.4.5

MODIFIED BITUMEN AND EMULSIONS .................................................................. 39

BITUMEN EMULSIONS ..............................................................................................................49 5.1

BACKGROUND ............................................................................................................. 49

5.2

DESCRIPTION ............................................................................................................... 49

5.3

MANUFACTURING OF EMULSIONS ........................................................................ 51

5.4

TYPES AND CLASSIFICATION OF EMULSIONS .................................................... 52

5.5

BITUMEN EMULSION PROPERTIES ......................................................................... 54

5.5.1

CHEMICAL PROPERTIES ............................................................................................ 54

5.5.2

BREAKING OF EMULSION ......................................................................................... 57

5.5.3

ADHESION ..................................................................................................................... 60

AGGREGATES ..............................................................................................................................61 6.1.1

INTRODUCTION............................................................................................................ 61

6.1.2

TYPES OF AGGREGATE .............................................................................................. 61

6.1.3

GENERAL ....................................................................................................................... 66

TEST METHODS AND EQUIPMENT .......................................................................................67 7.1

INTRODUCTION ........................................................................................................... 67 ix

AH GREYLING

December 2012


8

7.2

ADHESION TESTS ........................................................................................................ 67

7.2.1

STATIC IMMERSION TESTS ....................................................................................... 68

7.2.2

DYNAMIC IMMERSION TESTS .................................................................................. 68

7.2.3

CHEMICAL IMMERSION TESTS ................................................................................ 68

7.2.4

MECHANICAL AND IMMERSION TRAFFICKING TESTS...................................... 68

7.2.5

COATING TESTS ........................................................................................................... 69

7.2.6

ADSORPTION TESTS .................................................................................................... 69

7.2.7

IMPACT TEST ................................................................................................................ 69

7.2.8

PULL -OFF TESTS ......................................................................................................... 70

7.3

PROPOSED TEST METHOD ........................................................................................ 72

7.3.1

PATTI TEST .................................................................................................................... 72

7.3.2

INITIAL TEST METHOD LIMITATIONS & MODIFICATIONS ............................... 73

DEVELOPMENT OF THE BITUMEN BOND STRENGTH TEST ........................................74 8.1

INTRODUCTION ........................................................................................................... 74

8.2

UNIVERSITY OF STELLENBOSCH INVOLVEMENT ............................................. 74

8.3

MODIFICATION TO ORIGINAL PATTI TEST .......................................................... 76

8.3.1

INTRODUCTION............................................................................................................ 76

8.3.2

FILM THICKNESS ......................................................................................................... 76

8.3.3

LOADING RATE ............................................................................................................ 78

8.3.4

SUBSTRATE & SURFACE ROUGHNESS ................................................................... 78

8.4

TEST APPARATUS AND SETUP ................................................................................ 78

8.4.1

INTRODUCTION............................................................................................................ 78

8.4.2

STEP 1 – AGGREGATE PREPARATION ..................................................................... 79

8.4.3

STEP 2- AGGREGATE LAPPING ................................................................................. 80

8.4.4

STEP 3- CLEANING OF AGGREGATE ....................................................................... 81

8.4.5

STEP 4 – HEATING OF EMULSION ............................................................................ 82

8.4.6

STEP 5 HEATING OF AGGREGATE ........................................................................... 82

8.4.7

STEP 6-EMULSION ON AGGREGATE SAMPLE ....................................................... 82

8.4.8

STEP 7 – CURING OF SPECIMEN ............................................................................... 83

8.4.9

STEP 8 -APPLYING THE PULL-OUT STUB ............................................................... 84

8.4.10 STEP 9 -SAMPLE PREPARATION ............................................................................... 84 x AH GREYLING

December 2012


8.4.11 STEP 10 -TESTING ........................................................................................................ 84 8.5 9

GENERAL COMMENTS ............................................................................................... 87

EXPERIMENTAL SETUP ...........................................................................................................88 9.1

INTRODUCTION ........................................................................................................... 88

9.2

MATERIALS .................................................................................................................. 88

9.2.1

EMULSIONS ................................................................................................................... 88

9.2.2

AGGREGATES ............................................................................................................... 88

9.3

EXPECTED RESULTS .................................................................................................. 89

9.4

EXPERIMENTAL SETUP ............................................................................................. 89

9.5

TEST RESULTS- EMULSIONS .................................................................................... 90

9.5.1

CATIONIC 65% EMULSION (CRS65).......................................................................... 90

9.5.2

ANIONIC 60% EMULSION (SS60) ............................................................................... 93

9.5.3

CATIONIC 65% +3% LATEX MODIFIED EMULSION .............................................. 95

9.5.4

ANIONIC% +3% LATEX MODIFIED EMULSION ..................................................... 97

9.6 CONFIRMING SIGNIFICANT FACTORS INFLUENCING BOND STRENGTH DEVELOPMENT ........................................................................................................................ 99

10

11

9.6.1

LOADING RATE EXPERIMENT .................................................................................. 99

9.6.2

CURING CONDITIONS EXPERIMENT ..................................................................... 101

9.6.3

EMULSION TYPE EXPERIMENT .............................................................................. 102

9.7

CORRELATIONS OF BBS TEST RESULTS ............................................................. 108

9.8

INTER-LABORATORY EVALUATION OF BBS TEST METHOD ........................ 110

CONCLUSION & RECOMMENDATIONS .............................................................................113 10.1

CONCLUSION ............................................................................................................. 113

10.2

RECOMMENDATION FOR FUTURE STUDIES ...................................................... 114

BIBLIOGRAPHY ........................................................................................................................115

xi AH GREYLING

December 2012


TABLE OF FIGURES Figure 1-1- Study Tour Location .................................................................................................................. 3 Figure 3-1- Visual Representation of a Single and Double Chip and Spray Seal (South African National Road Agency, 2007) ................................................................................................................................... 17 Figure 3-2- Typical Design chart for a single seal using aggregate with ALD=9mm (South African National Road Agency, 2007) ..................................................................................................................... 19 Figure 3-3- Kearby Nomograph to determine asphalt cement application rate in chip seals (Kearby, (1953)) ........................................................................................................................................................ 22 Figure 3-4 – Schematic illustration of a single seal and the terms relevant to a single seal ....................... 25 Figure 4-1- Broad Chemical Composition of Bitumen (SABITA-M2, 2007) ............................................ 37 Figure 4-2- Typical effect of F-T wax on viscosity over temperature ........................................................ 41 Figure 4-3- Effect of elastomers on the rheological profile of bitumen (SABITA-M2, 2007)................... 45 Figure 4-4- Typical force-ductility curves for various modified binders (Asphalt Academy-TG1, 2007) . 47 Figure 4-5- Adhesion characteristics of Cationic latex modified emulsions (Asphalt Academy-TG1, 2007) .................................................................................................................................................................... 48 Figure 5-1- Micrograph of Bitumen Emulsion (James, 2006) .................................................................... 50 Figure 5-2- Colloidal Mill (Jaixing Mide Machinery Company, 2012) ..................................................... 51 Figure 5-3- Bitumen Emulsion Manufacturing Process (Asphalt Academy-TG2, 2009)........................... 51 Figure 5-4- Cationic Emulsifier Molecule (James, 2006) ........................................................................... 54 Figure 5-5- Anionic Emulsifier (Bickford, 2001) ....................................................................................... 55 Figure 5-6- Cationic Emulsifier (Bickford, 2001) ...................................................................................... 56 Figure 5-7- Possible stages in the setting of cationic emulsion (James, 2002) ........................................... 58 Figure 5-8- Typical stages in the breaking of emulsions (James, 2006) ..................................................... 59 Figure 5-9- Emulsion reaction with Acid Rock Types (Bickford, 2001) ................................................... 60 Figure 7-1- Vialit Pendulum test from the Shell Bitumen Handbook (Shell Bitumen, 2003) .................... 69 Figure 7-2- Cross section schematic of PATTI piston attached to a pull stub (Youtcheff & Aurilio, 1999) .................................................................................................................................................................... 71 Figure 7-3- Original PATTI assembly. ....................................................................................................... 72 Figure 8-1- PATTI Quantum Gold user interface ....................................................................................... 86 Figure 9-1- CRS65 Test Results on Granite & Tillite................................................................................. 91 Figure 9-2- SS60 Test Results on Granite & Tillite.................................................................................... 93 Figure 9-3- CRS65+3% Latex - Test Results on Granite & Tillite............................................................ 95 xii AH GREYLING

December 2012


Figure 9-4- SS60+3% Latex - Test Results on Granite & Tillite ................................................................ 97 Figure 9-5- Loading rate and pull-out tension are described by a power law model. (Miller, 2010) ....... 100 Figure 9-6- Loading rate and pull-out tension results US ......................................................................... 101 Figure 9-7- Pull-out tension values differ for various emulsion types at a range of curing intervals (Miller, 2010) ......................................................................................................................................................... 103 Figure 9-8- Pull-out tension values differ for various substrate types at a range of curing intervals (Miller, 2010) ......................................................................................................................................................... 103 Figure 9-9- Pull Out Tension Values for SS60& CRS65 on Granite & Tillite ......................................... 104 Figure 9-10- Pull Out Tension Values for SS60+3% Latex CRS65+3% Latex on Granite & Tillite ...... 105 Figure 9-11- Pull Out Tension Values for CRS65& CRS65+3% Latex on Granite & Tillite .................. 106 Figure 9-12- Pull Out Tension Values for SS60& SS60+3% Latex on Granite & Tillite ........................ 107 Figure 9-13- When sweep test results are compared to BBS test results, a potential BBS specification limit may be proposed at 850 kPa to define a specification target range (Miller,2010) ........................... 109 Figure 9-14- BBS test results compare well to DSR strain sweep results at two strain levels. ................ 110 Figure 9-15- Pull-out tension values differ for various emulsion types at a range of curing intervals. .... 111 Figure 9-16- Pull-out tension values differ for various substrate types at a range of curing intervals. .... 111

xiii AH GREYLING

December 2012


LIST OF TABLES Table 3-1- Typical Modes and Types of Distress on Surfaced Roads (CSIR-TRH6, 1985) ...................... 13 Table 3-2- Recommended Binders for different Traffic and Climatic conditions (South African National Road Agency, 2007) ................................................................................................................................... 32 Table 3-3- Proposed New Zealand Performance Grading System (Opus International Consultants, 2010) .................................................................................................................................................................... 34 Table 4-1- South African Penetration Grade Bitumen Requirements......................................................... 38 Table 4-2- Modified Binder Classification System (Asphalt Academy-TG1, 2007) ................................. 43 Table 5-1- South African Anionic Bitumen Emulsion Specification (SANS 309, 2004) ........................... 53 Table 5-2- South African Cationic Bitumen Emulsion Specification (SANS 548, 2003) .......................... 53 Table 5-3- Chemistry of Bitumen Emulsifiers (James, 2002) .................................................................... 55 Table 5-4- Typical Emulsion Recipes (James, 2006) ................................................................................. 57 Table 6-1- Characteristic combinations of rock forming materials in rocks (Weinert, 1980) .................... 63 Table 6-2- Rock Classification System (Weinert, 1980) ............................................................................ 64 Table 9-1- Experimental Setup and Test Sets completed. .......................................................................... 90 Table 9-2- Covariance Analysis of CRS Test Set 5 on Granite .................................................................. 92 Table 9-3- Covariance Analysis of SS60 Test Set 8 on Granite ................................................................. 94 Table 9-4- Covariance Analysis of SS60 Test Set 8 on Tillite ................................................................... 94 Table 9-5- Covariance Analysis of CRS65+3% Latex Test Set 7 on Granite ............................................ 96 Table 9-6- Covariance Analysis of CRS65+3% Latex Test Set 7 on Tillite............................................... 96 Table 9-7- Covariance Analysis of SS60+3% Latex Test Set 10 on Granite ............................................. 98 Table 9-8- Covariance Analysis of SS60+3% Latex Test Set 10 on Tillite ................................................ 98 Table 9-9- Analysis of variance for the factor screening experiment. (Miller, 2010) ................................ 99 Table 9-10-Analysis of variance for the curing conditions experiment. (Miller, 2010) ........................... 102 Table 9-11-BBS test results for materials tested at the University of Stellenbosch. ................................ 112

xiv AH GREYLING

December 2012


LIST OF PHOTOS Photo 2-1- Restored Fowler Roller (www.TractionTime.co.uk) ................................................................ 10 Photo 2-2- Early Bitumen Spray Operations (SABITA, 2009) .................................................................. 10 Photo 3-1- Areas showing ravelling and stone loss. ................................................................................... 15 Photo 3-2- Area showing signs of bleeding ................................................................................................ 16 Photo 8-1- Modified pull-out stubs with metal support blocks (Kanitpong & Bahia, 2003) ...................... 76 Photo 8-2- Modified pull-out stub with aluminum frame supports. ........................................................... 77 Photo 8-3- Modified BBS test pull-out stubs. ............................................................................................. 77 Photo 8-4- Rock Slab Saw (Miller, 2010)................................................................................................... 79 Photo 8-5- Rock Slab Saw and Screw Driven Mechanism. (Miller, 2010) ................................................ 79 Photo 8-6- Lapping Machine (Constable, 2009) ......................................................................................... 80 Photo 8-7- Ceramic Grit 280 (Constable, 2009) ......................................................................................... 81 Photo 8-8- Ultrasonic cleaner ..................................................................................................................... 81 Photo 8-9- Granite and Tillite prepared aggregate samples ........................................................................ 82 Photo 8-10- Bitumen emulsion poured into silicone moulds ...................................................................... 83 Photo 8-11- Environmental Chamber ......................................................................................................... 83 Photo 8-12- Core Aggregate samples in Forced Draft Oven (Constable, 2009)......................................... 84 Photo 8-13- PATTI Quantum Gold with testing assembly (Constable, 2009) ........................................... 85 Photo 8-14- Top View of PATTI Quantum Gold ....................................................................................... 85 Photo 8-15- Loading Rate dial .................................................................................................................... 86

xv AH GREYLING

December 2012


LIST OF ADDENDUMS ADDENDUM 1- FINAL DRAFT OF BBS TEST PROCEDURE - LATER ACCEPTED AS (AASHTO TP-91 , 2011) ............................................................................................................................................ 119 ADDENDUM 2- UWM STUDY TOUR FEEDBACK REPORT ........................................................... 120 ADDENDUM 3- MINUTES OF STUDY GROUP CONFERENCE CALLS AND REPORTS COMPLETED DURING THIS PHASE................................................................................................... 121 ADDENDUM 4- BBS TEST RESULTS OF TESTS COMPLETED IN 2010 ........................................ 122 ADDENDUM 5- PAPER PRESENTED AT 2ND INTERNATIONAL SPRAY SEALING CONFERENCE, AUCKLAND, AUSTRALIA, 2010 ............................................................................. 123 ADDENDUM 6 – PAPER PRESENTED AT 10TH CONFERENCE ON ASPHALT PAVEMENT FOR SOUTHERN AFRICA, 2011.................................................................................................................... 124

xvi AH GREYLING

December 2012


GLOSSARY/ACRONYMS ALD

Average Least Dimension'

COV

Covariance

CRS65

Cationic Rapid Set Emulsion with 65% net bitumen content

CSIR

Council for Scientific and Industrial Research

DSR

Dynamic Shear Rheometer

EVA

Ethylene-Vinyl-Acetate

FHWA

Federal Highway Administration

HMA

Hot Mix Asphalt

PATTI

Pneumatic adhesion tensile testing instrument

PG

Performance Grading

PQG

PATTI Quantum Gold

PSV

Polished Stone Value

SABITA

Southern African Bitumen Association

SABS

South African Bureau of Standards

SBR

Styrene-Butadiene-Rubber

SBS

Styrene-Butadiene-Styrene

SHRP

Strategic Highway Research Programme

SS60

Stable Grade Anionic Emulsion with 60% net bitumen content

TG

Technical Guideline

TMH

Technical Methods for Highways

UAI

University of Ancona-Italy

US

University of Stellenbosch

USA

United States of America

UWM

University of Wisconsin - Madison

xvii AH GREYLING

December 2012


1

INTRODUCTION

1.1 BACKGROUND Chip and spray surfacing seals have been widely used in South Africa as the preferred surface treatment for rural roads. The design of these seals has also experienced renewed interest and continuous development in several other countries over the past two decades. In South Africa seals are continually used as increasing attention is given to the periodic maintenance of existing surfaced roads. There is also a significant upward trend in the use of surfacing seals in North America as the need to develop more energy and resource efficient surfacing options becomes a priority.

Despite this growing use of surface seals, the seal design and especially the selection of binder type and grade does not always follow scientific processes. Seals are often designed based on client preference, previous experience, material availability and industry trends. With an ageing road network and limited funding to ensure timeous maintenance, the focus in South Africa is moving towards more stringent and scientific design processes.

The USA and Europe, forced by increasing traffic volumes and heightened performance demands, are moving towards performance- based specification to account for deficiencies in their current surface seal design methods (Opus International Consultants, 2010). One of the major failure mechanisms of surface seals is ravelling which takes place when the binder and the aggregate bond becomes weak and are broken by the forces generated by traffic. This leads to loose aggregate on the road, which in turn, leads to bare surface patches and broken windscreens. To prevent and address this as part of the development of performance-based specifications, the need for a simple and inexpensive technique for evaluating bitumen and bitumen emulsion bond strength development over time, as well as binder-aggregate compatibility, was identified.

Although various tests exist for investigating adhesion between bituminous emulsions and aggregate chips (European Commitee for Standarization, 1999) most of the tests do not deliver the level of information required by the performance-based specifications.

The Bitumen Bond Strength (BBS) test method was therefore

developed with the aim to address some of the limitations encountered in evaluating bond strength between binders and aggregates.

1 AH GREYLING

December 2012


The BBS test (AASHTO TP-91 , 2011) was developed by the University of Wisconsin – Madison (UWM) in partnership with the University of Ancona – Italy (UAI) and the University of Stellenbosch – South Africa (US) specifically for evaluating bond strength between aggregates and hot applied binders and emulsions, respectively.

US became involved in BBS test efforts in 2008 to assist in the development and practical evaluation of the BBS test method. UAI contributed significantly to the development of the test apparatus through their work in conjunction with UWM. Due to time and resources available, the involvement of the US was limited to various discussion sessions, a study tour, the evaluation of the BBS test, and conducting a series of control tests.

The developers of the BBS test foresee that the test method will be able to quantify and characterize bond strength development, adhesive properties and aggregate-binder compatibility, thereby improving the effectiveness of surface seal designs (Miller, 2010).

1.2 PROJECT DEFINITION AND OBJECTIVES 1.2.1 PROJECT DEFINITION The project title can be defined as follows: Development of a Standard Test Method for Determining the Bitumen Bond Strength of Emulsions- A South African Perspective.

1.2.2 RESEARCH OBJECTIVES The research objectives of this study are to address the following identified outcomes:

1. Complete a detailed literature review to identify the factors that significantly affect the bond strength development between bituminous binders and aggregates. 2. Confirm the need for a test method that will be able to quantify and characterize bond strength development, adhesive properties and aggregate-binder compatibility. 3. Describe the writer’s assistance in the development of a test method to quantify bond strength development in bitumen and emulsified bitumen binders. 2 AH GREYLING

December 2012


4. Evaluate the practicality and repeatability of the developed Bitumen Bond Strength (BBS) Test procedure. 5. Evaluate the bond strength development of modified and unmodified bitumen emulsions on tillite and granite aggregates using the Bitumen Bond Strength Test method. 6. Correlate the results achieved at the University of Stellenbosch with results from the University of Wisconsin-Madison

1.3 STUDY LOCATION The study took place at the University of Stellenbosch in the Western Province of South Africa and at the University of Wisconsin-Madison in the state of Wisconsin in the United States of America. Figure 1-1 below shows locations in relation to a world map.

Figure 1-1- Study Tour Location

3 AH GREYLING

December 2012


1.4 METHODOLOGY The study was divided into three distinct phases: 

The first phase consisted of an extensive literature study and review on various subjects relating to surfacing seals, bitumen, aggregate and their behaviour. It investigated the factors influencing bitumen and aggregate adhesion as well as the current test methods available to evaluate adhesion. The need and initial development of the Bitumen Bond Strength test is also addressed.



The second phase details the physical processes and actions followed to assist in the development of the Bitumen Bond Strength test. It includes the evaluation of the BBS Test procedure during its development phase by completing a number of trail tests. This phase also investigates the final procedure by completing a detailed experimental setup and an extensive set of tests.



The third and final stage is the evaluation of the study results and includes a discussion on required future research.

1.5 THESIS OUTLINE The thesis can be divided into five sections:

Chapter 1 gives a brief introduction of the proposed study; it defines the project definition and objectives and discusses the methodology followed to complete the study. Chapter 2 consists of a brief introduction on the history of roads and the use of bitumen surfacing seals.

Chapters 3 to 6 focusses on road surfacing seals, their function , behaviour and the material used in seals. It discusses the design methods of seals, the modes of distress and factors that typically influence the life cycle of roads. It also looks at bitumen, bitumen emulsion and aggregates and these materials’ properties and specification systems.

Chapter 7 discusses the current test methods available for testing adhesion and evaluates the PATTI Test as a possible new alternative. 4 AH GREYLING

December 2012


Chapter 8 & 9 discusses the involvement of the University of Stellenbosch, the development of the Bitumen Bond Strength Test as well as the test results. In addition, factors that influence the bitumen bond strength development are confirmed.

Chapter 10 concludes the report by discussing how the research objects was met and proposes some recommendations for future studies.

5 AH GREYLING

December 2012


LITERATURE STUDY: BITUMEN &AGGREGATE ADHESION 2

AN INTRODUCTION TO ROADS AND ROAD SURFACING SEALS

2.1 A BRIEF HISTORY OF ROADS Most people give very little thought to the roads we drive on every day. We tend to take them for granted, at least until they are in a bad condition or closed for any reason. From the earliest times, one of the strongest indicators of a society's level of development has been its road system or lack of one. Increasing populations and the introduction of towns and cities brought with it the need for communication and trade between those growing population centres. A road built in Egypt by the Pharaoh Cheops around 2500 BC is believed to be the earliest paved road on record. It consisted of a 1,000 meters long and 20 meters wide construction road that led to the site of the Great Pyramid.

Various other ancient roads were established by different civilizations and their armies but the Romans where, without a doubt the champion road builders of them all. Roman roads were masterpieces of road construction, and was designed and built with great engineering skill. At their peak, the Roman Empire maintained around 85,000 km of roads that covered almost all of Europe. Roman roads consisted of a graded soil foundation topped by four layers: a bedding of sand or mortar; rows of large, flat stones; a thin layer of gravel mixed with lime; and a thin surface of flint-like lava. Many of their original roads are still in use today, although they have been resurfaced numerous times.

Thomas Telford and John Loudon McAdam, two well-known British Engineers, initiated the modern highway construction technology in 18th century Britain. Telford, originally a stonemason, came up with a system of road building which required digging a trench, installing a foundation of heavy rock, and then surfacing with a 150mm layer of gravel. During construction, the centre of the road was raised, producing a crown that allowed water to drain off. Telfords’ system was faster and less expensive than the Romans' method but it was still costly and required frequent resurfacing with gravel.

On the other hand, McAdam's system was based on the principle that a well-drained road made of suitable material does not need the stone foundation of Telford's system, but could be built directly on the subsoil. McAdam firstly placed a closely compacted 250mm- to 300mm layer of stone which had been broken to 25mm 6 AH GREYLING

December 2012


in diameter, and which was raised in the centre to facilitate drainage. This was followed by a layer of finer grained stone that was cemented by the setting of fines of the graded stone. This process was completed in stages and it allowed the road's traffic to compact each stage.

The greatest advantages to McAdam's system were the speed of construction and low cost, and it was generally adopted throughout Europe. However, it was the lack of a firm foundation for the roadbed that was to prove the ultimate undoing of macadam roads with the advent of heavy motor vehicles, especially trucks. For that reason, on roads that had to support heavy loads, Telford's system of construction became the standard (Sponholtz, 2012).

2.2 HISTORY OF BITUMEN SURFACING SEALS 2.2.1 REST OF THE WORLD The history of modern road construction and paving materials dates back to the days of the late 1800's when it became widespread in all major cities around the world. The chief paving materials of the day consisted of loose cobble stones and blocks in addition to miscellaneous wooden bricks, blocks, crushed rocks and pebbles, and at times some naturally occurring tar and cement concrete surfacing materials were also used.

As discussed in the previous section the design of John McAdam provided the most cost effective and widely used paving options at the time. Its open graded surface was unfortunately not very suitable to sustain a combination of heavy traffic and moisture and resulted in slippery and muddy road surfaces. At the start of the 20th century, roads were mostly inadequate to cater for the everyday traffic demands which were going to be challenged increasingly by the newly invented motorcar. As time went by and the motor car technology developed and vehicle velocities increased, the friction between the road surfaces and the car tyres became a critical road safety issue, especially when subjected to stresses encountered during constant acceleration, cornering and braking conditions.

In addition to numerous pavement failures experienced at the time it was also made apparent that much more durable and tougher road construction materials were called for. The answer at the time was relying in the ongoing research which was being conducted to find better, cheaper and stronger road pavement materials and processes. There was a need for a road surface that could withstand the forces of everyday road traffic. Roads made from asphalt and concrete seemed to offer the most promise. 7 AH GREYLING

December 2012


Asphalt is generally a mixture of bitumen and aggregates consisting of crushed rock whereas concrete is a mixture of cement and rock. Asphalt pathways were first paved in France in the early 1800's, but the method was not perfected until around 1829. The first use of roads constructed using asphalt and bitumen took place at around 1825, when asphalt blocks were first put down as a pavement on French streets in Paris, and the first successful great asphalt and bitumen (tar) paving application was first laid down in 1858. The first successful paved cement concrete sidewalk in turn was built in Inverness in Scotland in 1865, neither technology however progressed greatly without the added pressures which the motorcar posed on the roads infrastructure in the coming years (Ecopave Australia, 2009).

The history and advancement of the modern asphalt/bitumen road, as we see on our roads today, came initially from the United States. They had vast natural occurring bitumen deposits from which to draw from and the local engineers were forced to analyse the principles behind the behaviour of this promising new road engineering material.

The first real initiative came in the late 1860's, from the research work conducted by a Belgian immigrant Edward de Smedt from Columbia University in New York City. Mr De Smedt conducted his first field testing in New Jersey in 1870 and by 1872 was already producing the equivalent of modern day continuously graded road asphalt. The first practical application was followed by the laying of asphalt in Battery Park and on Fifth Avenue in New York USA in 1872 (Sponholtz, 2012).

One of the great convenient coincidences in the history of asphalt/bitumen roads was the development of the motor car which by pure chance ran on petrol. Petrol, at that time, was just another petroleum by-product which was distilled from kerosene which in turn was derived from petroleum by the fast emerging and growing petroleum refining industry. Another useful by-product that was derived in the process of refining crude oil was bitumen (Ecopave Australia, 2009).

8 AH GREYLING

December 2012


2.2.2 SOUTH AFRICA Paved Roads in South Africa can be described as fairly recent as they have been present for just over 100 years. The first new asphalt roads in South Africa that were laid in Cape Town soon after the turn of the 20th century. A Mr John Whitford Griffits arrived in Cape Town from Wales in 1902 and he started a road construction company called the Union Granolithic and Asphalt Company in 1908. This company surfaced Church, Longmarket, Shortmarket, Hout and Castle streets in 1910 using bitumen imported from Switzerland.

After a break in activity during the First World War, the main road system in and around Cape Town was progressively reconstructed with asphalt surfaces. Shell Oil imported Trinidad Lake Asphalt as binder and, as the City Council of Cape Town had ample funds at the time; specifications based on sand seals to penetration layers using granite aggregates became the order of the day (SABITA, 2009).

The plant used by these early road builders requires some comment. Mr Wolton Gray came to South Africa in 1924 to sell steam rollers for John Fowler of Leeds, England. The steam rolling business was slow to take off so Mr Wolton launched Fowler Tar Spraying with two fellow steam men, Mr Lowman and Mr Fishwick. They started off by hiring out the Fowler rollers and eventually went on to undertake road construction contracts. Some of their early work included the roads and wharves of Cape Town Docks for the S. A. Railways & Harbours.

The Fowler roller was unique in that it had a tar kettle under the boiler and a chain driven tar pump feeding a 2.750m (9ft) spray bar at the rear end. It came equipped with a drawn hopper which spread chips onto the freshly sprayed binder. This method is in all probability the forerunner of the Chip Seal currently used throughout South Africa.

Photo 2-1 on the next page shows a restored Fowler steam roller and Photo 2.2 shows the manual application of a bituminous binder (SABITA, 2009).

9 AH GREYLING

December 2012


Photo 2-1- Restored Fowler Roller (www.TractionTime.co.uk)

Photo 2-2- Early Bitumen Spray Operations (SABITA, 2009)

10 AH GREYLING

December 2012


3

ROAD SURFACING SEALS

3.1 INTRODUCTION The previous section gave a very brief overview of the history and background of bituminous surfacing treatments used in road construction. Asphalt, Chip and Spray and variations of the two are currently the most commonly used surfacing seals in South Africa.

Asphalt is a blend of graded aggregate, bitumen and a filler material that is typically mixed in a designated plant using a specific mix design and mix procedure. It is hot or warm applied on the road using an asphalt paver after which it is rolled to a specified density. This is the most durable but also the most expensive bituminous surfacing type and it is generally only provided on heavily trafficked rural roads, freeways, urban streets and at intersections. Asphalt and the behaviour of asphalt do not form part of this study, but it is however envisaged that some of the outcomes may be relevant for use in the asphalt industry.

The size of South Africa and the location and distribution of our road network in relation to the large centres, as well as the relatively low traffic flow on some of our roads, has led to the extensive use of chip and spray seals. A chip and spray seal typically consist of a layer of bituminous binder sprayed directly on the road followed by an even sized or graded aggregate mix that is spread over the binder before being rolled and opened to traffic (South African National Road Agency, 2007).

It is used in various combinations of binders and aggregate sizes applied in different layer combinations. It is the most used seal and re-seal method in the rural areas of South Africa. It is also the only type of surfacing seal that will be further investigated as part of study and all further references to surfacing seal and seals will include only chip and spray seals unless otherwise stated.

3.2 SURFACING SEALS FUNCTION The Technical Recommendation for Highways Number 3 (TRH3) is the South African authority on surfacing seal design and is currently the proposed method to be used for chip and spray seal designs. It defines a good road as a structurally sound surface that has a durable, waterproof, skid-resistant and all weather dust-free surfacing to ensure that it perform optimally (functionally and structurally). These characteristics are required to

11 AH GREYLING

December 2012


provide the road user with an acceptable level of service and to protect the structural layers of the pavement from the forces of traffic as well as from the effects of the environment In South Africa chip and spray seals and slurries are commonly used for new construction and for the resealing of existing bituminous roads. This is mainly because they are relatively inexpensive and simple to construct. They have proved to be successful on highways, rural roads and urban streets, under light and heavy traffic conditions.

The main functions of a surfacing seal as per TRH3 can therefore be summarised as follows: o

Provide a waterproof cover to the underlying crushed stone and natural material pavements;

o

Provide a safe all-weather, dust-free riding surface for traffic with adequate skid resistance;

o

Protect the underlying layer from the abrasive and destructive forces of traffic and the environment.

Most surfacing seals are relatively thin and have no load distribution properties. The seal itself should however have enough strength to accommodate the horizontal and vertical stresses induced by traffic (South African National Road Agency, 2007).

3.3 SURFACING SEAL FAILURE MODES AND MECHANISMS 3.3.1 BACKGROUND The functional requirements of any road are speed, comfort and safety. These functions are almost always related with relationships that are not easily quantifiable or measurable. For this reason various functional features have been defined. These are features that can be measured and for which a standard can be set to ensure suitable levels of comfort, safety and speed. For surfaced roads these are generally the riding quality, skid resistance and surface drainage of the road.

3.3.2 MODES OF DISTRESS Distress is the visible manifestation at the road pavement surface of the deterioration of the condition of the pavement with respect to either the serviceability or the structural capacity. Part of a comfortable and safe road is a road with limited signs of distress that will negatively influence the riding quality, skid resistance or surface drainage of a road. There are four major ways in which distress occurs (CSIR-TRH6, 1985). These are called the modes of distress and they can be summarised as follow: 12 AH GREYLING

December 2012




Deformation – This is the development of a change in the profile of the surface of the pavement which leads to unevenness and reduce riding quality.



Cracking – This refers to cracks that appear in the surface of the pavement.



Disintegration of the surfacing - This is the break-up and loss of surfacing. It can occur through spalling of the material at cracks or through the crumbling and loss of material at the surface due to the abrasive action of traffic.



Smoothing of Surface Texture - This is the loss of surface texture which leads to a loss of skid resistance.

Each of these modes of distress can occur in one of several different typical manifestations. These are called the various types of distress. Table 3-1 below summarise the modes and types of distress found on surfaced roads: Table 3-1- Typical Modes and Types of Distress on Surfaced Roads (CSIR-TRH6, 1985) MODE OF DISTRESS Deformation

TYPE OF DISTRESS Depressions Mounds Ruts Ridges Displacements Corrugations Undulations

Cracking

Transverse cracks Longitudinal cracks Block Cracks Map Cracks Crocodile Cracks Parabolic Cracks Star Cracks Meandering Cracks Multiple Cracks

Disintegration of Surfacing

Ravelling Potholes Edge Breaks Patches

Smoothing of surface texture

Bleeding &Polishing

13 AH GREYLING

December 2012


When looking at these modes and types of failure and distress it is extremely important to understand and differentiate between what is in essence structural distress and what is surfacing distress. Structural distress refers to distress that affects the structural integrity and the load-bearing capacity of a pavement in either origin or consequent. Surface distress is that distress that affects only the surface texture and behaviour. (CSIR-TRH6, 1985)

As part of this study structural distresses will not be investigated further. The following sections will briefly discuss the typical surface distresses experienced under each mode of distress. The importance, effect and influence of the seal design, binder and aggregate on these distresses will be pointed out in subsequent sections.

3.3.3 MODES OF SURFACE DISTRESS 3.3.3.1 DEFORMATION Deformation in a road is almost always related to structural issues. No distresses related only to thin chip and spray surfacing’s relevant and it will therefore not be discussed further. 3.3.3.2 CRACKING Cracking of the surfacing is either due to reflective cracking from problems in the underlying structure or due to the ageing of a seal. When a seal ages it becomes dry and brittle, losing much of its elasticity. This leads to various types of localised cracks on the surface. Surface cracks may lead to water ingress which in turn weakens the support layers and lead to disintegration of the surface. 3.3.3.3 DISINTEGRATION OF THE SURFACING Ravelling and stone loss is one of the most important surfacing seal related distresses. Ravelling occurs when aggregate “ravel” from the surface due to loss of adhesion between the binder and aggregate. This forms depressions which may fill with moisture which may pose problems. Loose aggregate on the road may also lead to vehicle and windscreen damage. Ravelling can be due to any of the various reasons that influence the bitumen binder and aggregate adhesion. These reasons will be discussed in more detail further on in this study. Photo 3-1 below shows a typical ravelled strip on a chip seal.

14 AH GREYLING

December 2012


Photo 3-1- Areas showing ravelling and stone loss.

3.3.3.4 BLEEDING AND SMOOTHING OF SURFACE Bleeding and the loss of surface texture are further surface seal specific defects. Bleeding takes place when the voids within the aggregate matrix, that should accommodate the bitumen binder, are reduced by either aggregate that embeds or punches into the base course or the rotation of stone to their least dimension. This decrease in voids, coupled with high temperatures that lead to decreased bitumen viscosity, causes the road to “bleed” bitumen.

The liquid bitumen migrating to the surface creates a sticky and shiny, black surface. The result can mean a loss of surface texture on the pavement. Photo 3-2 below shows a section of road showing signs of bleeding. Note the darker strip in the right wheel track of the left lane.

15 AH GREYLING

December 2012


Photo 3-2- Area showing signs of bleeding

A second cause of the smoothing of the surface may be polished aggregate. This happens when the bitumen has been worn away from the pavement surface by traffic, and the surface aggregate has lost macro texture, having been smoothed and rounded. This happens when the aggregate has insufficient resistance against polishing as defined by the Polished Stone Value (PSV) and measured according to SABS Method SM84818.

The degree of possible polishing depends on the aggregates mineral composition and crystal structure. These properties will be discussed further in Chapter 6 of this report.

16 AH GREYLING

December 2012


3.4 SURFACING SEAL DESIGN 3.4.1 INTRODUCTION The TRH3- South African National Road Agency, 2007 defines a chip and spray seal as follows: “In its simplest form a chip and spray seal consists of a coat of bituminous binder sprayed onto the road surface which is then covered with a layer of aggregate (stone or sand). The aggregate cover is applied immediately after the binder has been sprayed and then rolled to ensure close contact and thus good adhesion between the aggregate and the binder film. Rolling initiates the process of orientating the particles into a mosaic pattern and working the binder into the voids between the aggregate particles. The process is completed by the action of traffic, so that finally a dense and relatively impermeable pavement surfacing is obtained.”

This method is used in various combinations to construct single seals, double seals, Cape Seals, slurry seals and sand seals. Further alternatives include inverted double seals, geotextile seals, split seals, graded aggregate seals and choked seals. Figure 3-1 below, extracted from the TRH3, shows visuals of the two most common used seal types

Figure 3-1- Visual Representation of a Single and Double Chip and Spray Seal (South African National Road Agency, 2007)

The very early practitioners of chip seals appear to have used a purely empirical approach to their seal designs. Sealing a pavement was considered then, as it is now in many circles, an art. Scientific developments in the field have yielded a few relevant design methods that vary across the world. The following section will aim to describe the essence and the philosophy of the majority of these methods.

17 AH GREYLING

December 2012


3.4.2 SOUTH AFRICAN METHOD, TRH3 The TRH3 was published by the South African National Road Agency in 2007 as the authority on South African chip and spray seal designs. The TRH3’s design process is based on a number of primary input parameters such as traffic volume, preferred new texture depth and surface hardness. Once initial results are achieved using the primary inputs, practical adjustments are made for differences in the climate, gradients, existing texture, applications temperatures, preferred aggregate spread rate, and the use of modified binders. This method may be seen as a hybrid of the United Kingdom and Australian design methodologies. It is however strongly influenced by extensive South African seal experience.

The selection of type of surfacing is usually made between single seals, double seals, Cape seals, and sand seals. The seal type decision is primarily based on the traffic level and pavement condition. The TRH3 method also measures and evaluates the surface hardness and expected penetration of aggregate by using a ball penetration test, corrected for temperature.

The grade and type of binder is selected based on expected traffic level, expected operational road surface temperature, climatic region, and aggregate condition. The required rate of binder spread is determined by using tables and charts that incorporate traffic level, expected aggregate embedment and final required texture depth for different aggregate Average Least Dimension (ALD) values.

Figure 3-2 shows the typical charts that will be used for a 13.2mm single seal. Various practical adjustments are then made to ensure the application is case specific and relevant to the specific conditions. Note the red minimum application line. This is the minimum binder required that can be practically be sprayed or that is required to accommodate and bind with the aggregate.

The aggregate spread rate varies according to the purpose of the seal and the shape and flakiness of the aggregates. The TRH3 proposes that final design spread rates should be determined on site by spreading the preferred matrix of aggregates by hand. (South African National Road Agency, 2007)

18 AH GREYLING

December 2012


Figure 3-2- Typical Design chart for a single seal using aggregate with ALD=9mm (South African National Road Agency, 2007)

19 AH GREYLING

December 2012


3.4.3 REST OF THE WORLD 3.4.3.1 INTRODUCTION The TRH3 is one of the newest methods in the world and it takes a lot of inspiration from international design methods. The section below will aim to give an overview of the current and historic methods used in the rest of the world. 3.4.3.2 HANSON METHOD In 1935 New Zealand’s F.M. Hanson was the first to present a scientific approach to the design and construction of chip seals in his paper “The bituminous surface treatment of rural highways” (Hanson, 1935) . The principles presented by Hanson initiated a practical approach to chip sealing that remained unchallenged for the next fifty years. His design method was developed primarily for bitumen and in particular cutback bitumen and it was based on the Average Least Dimension (ALD) of the aggregate spread on the road.

Hanson calculated the ALD by manually calipering a representative aggregate sample to obtain the smallest value for ALD that represents the typical thickness of the rolled aggregate layer. He observed that when aggregate is dropped from a chip spreader onto a bituminous binder, the void remaining between the aggregate particles is approximately 50% of the ALD. (This is true in most cases for a cubical type stone.) He theorized that when the layer is rolled, this value is reduced to 30% and it is further reduced to 20% when the aggregate is embedded or compacted by traffic.

His design method involved the calculation of bituminous binder and aggregate spread rates to be applied to fill a certain percentage of the voids between aggregate particles. Hanson specified the percentage of the void space to be filled by residual binder to be between 60% and 75%, depending on the type of aggregate and traffic level (Towler & Dawson, 2008). 3.4.3.3 MCLEOD METHOD Throughout the 1960’s, Mr Norman McLeod developed a design procedure based partially on Hanson’s previous work for use in the USA (McLeod, 1969). Mr McLeod’s design determines the aggregate application rate based on gradation, specific gravity, shape, and a wastage factor. He provided a correction factor based on the fraction of voids.

20 AH GREYLING

December 2012


The binder application rate depends on the aggregate gradation, absorption and shape, traffic volume, existing pavement condition, and the residual bitumen content of the binder. It should be noted that this method was developed primarily for use with emulsion binders. McLeod made it apparent that correction factors for the quantity of binder lost by absorption of aggregate and texture of existing surface are recommended. McLeod’s work also gives guidelines on the appropriate type and grade of bitumen for the selected aggregate and surface temperature at time of application (McLeod, 1969).

The United States of America’s Asphalt Emulsion Manufacturers Association and the Asphalt Institute have adapted this method in the form of recommendations for binder types and grades for various aggregate gradations, and correction factors to the binder application rate based on existing surface condition.

3.4.3.4 KEARBY METHOD In 1953, Mr J.P. Kearby, an engineer with the Texas Highway Department in the USA, made one of the first efforts at designing chip seal material application rates in the United States. Kearby’s design philosophy was that “computations alone cannot produce satisfactory results and that certain existing field conditions require visual inspection and the use of judgment in the choice of quantities of asphalt and aggregate” (Kearby, (1953)). This is still relevant and a very good design philosophy to follow.

Kearby developed a method to determine the amounts and types of bitumen and aggregate rates for chip seals. Kearby’s work resulted in the development of a nomograph, shown below as Figure 3-3. It developed provided a bitumen application rate in volume per area. It required input data in the form of average seal thickness, % aggregate embedment, and % voids. The design methodology requires the knowledge of some physical characteristics of the aggregate, such as unit weight, bulk specific gravity, and quantity of aggregate needed to cover a specific area of roadway (Gransburg, et al., 2010).

21 AH GREYLING

December 2012


Figure 3-3- Kearby Nomograph to determine asphalt cement application rate in chip seals (Kearby, (1953)) Kearby suggested that when surface treatments are applied over existing chip sealed or tightly bonded hard base courses, the percentage of embedment should be increased for hard aggregates and reduced for soft aggregates (Gransburg, et al., 2010) .

3.4.3.5 MODIFIED KEARBY METHOD (TEXAS) In 1974, Epps and associates proposed a change to the design monograph developed by Kearby to make it suitable for the use of synthetic aggregates in chip seals (Epps, et al., 1974). They further developed correction factors that were based on experiences and combinations that worked well in practice (Epps, et al., 1980) .

The developed binder application rate correction factors corresponded to traffic level and surface condition. Epps also suggested that consideration be given to varying the bitumen application rate both longitudinally and transversely, as reflected by the pavement surface condition. Since that time, this design approach has 22 AH GREYLING

December 2012


been labelled as the modified Kearby method by both practitioners and researchers. Since the publication of that design procedure, the Texas Department of Transportation’s Brownwood District has expanded on the bitumen application correction factors to include adjustments for truck traffic and existing surface condition.

The modified Kearby method also recommends a laboratory “board test” method to find the quantity of aggregate needed to cover a specific area of roadway. The board test is performed by placing an adequate number of chips on the required area (Gransburg, et al., 2010).

3.4.4 ROAD NOTE 39-UK The United Kingdom’s Transport Research Laboratory has published several editions of a comprehensive design procedure for “surface dressing” roads in the United Kingdom. The technology that makes this design procedure so advanced is the extensive use of a computer design program based on decision trees. Known as Road Note 39, this design procedure is highly advanced and uses a multitude of input parameters. Traffic level, road hardness, surface conditions, and site geometry are all critical input factors. Skid-resistance requirements and likely weather conditions are secondary inputs into the program (Roberts & Nicolls, 2008). This procedure includes the following five steps: 1. Selection of the type of dressing—the selection of surface dressing (Chip Seal) is made from five treatments: single seal, texture slurry plus a single seal, racked-in seal, double seal, and sandwich seal. 2. Selection of binder—Binders are selected from either emulsion or cutback bitumen, specified based on viscosity. Modified binders such as polymer-modified binders are also recommended if their need and additional cost can be justified. 3. Selection of aggregate—The nominal size of aggregate is selected based on traffic and hardness of existing surface. Specified are 20-, 14-, 10-, 6-, and 3-mm nominal-size aggregates. 4. Binder spread rate—The required rate of binder spread depends on the size and shape of aggregates, nature of existing road surface, and degree of embedment of aggregate by traffic. The rate of binder spread should not vary by more than 10% from the target figure. 5. Rate of aggregate spread—The aggregate spread rate is determined based on a “tray test” and depends on the size, shape, and relative density of the aggregate. The basic inputs into the decision trees include selection of the type of treatment and selection of grade and type of binder based on traffic and construction season. 23 AH GREYLING

December 2012


The aggregate type and size are selected based on skid and friction requirements, likely weather conditions, and hardness of existing surface. The resulting design application rate of binder is determined by the size and shape of aggregates, nature of existing road surface, and degree of embedment of aggregate by traffic. The resulting design application rate of aggregate spread rate depends on the size, shape, and relative density of the aggregate (Gransburg, et al., 2010).

3.4.5 AUSTROADS SPRAYED SEAL DESIGN METHOD The 2004 Austroads published a Sprayed Seal Design Manual that provide a performance-based design method that uses an extensive list of input parameters for determining aggregate and binder application rates. Aggregate angularity, traffic volume, road geometry, ALD of aggregate, aggregate absorption, pavement absorption, and texture depth are the input variables for this method.

The main assumption of this design model is based on the original Hanson philosophy. It assumes that the aggregate in a seal is orientated approximately one layer thick and contains a percentage of air voids. Thus, filling a percentage of the voids with binder determines the binder application rate. The minimum binder application rate is determined by the percentage of voids to be filled, the total available voids, and the thickness of the seal.

The first step in the Austroads procedure is to determine a basic voids factor. Adjustments for aggregate characteristics and anticipated traffic levels are added to derive a design voids factor. That factor is then multiplied by the ALD of the aggregate to determine the basic binder application rate. This base binder application rate is then adjusted with allowances to cater to the texture and absorption of the pavement surface, the aggregate properties and road geometry. Further adjustments are made for designing as a reseal, but adjustments for surface texture and embedment are not performed (Austroads, 2004).

24 AH GREYLING

December 2012


3.5 GENERAL Although aggregate and binder adhesion is probably one of the most important requirement of any chip seal it is interesting to note that it is not a critical input parameter in any of the design methods. Some of the design methods propose test to ensure binder/aggregate compatibility but it is generally assumed that the selection of the binder and aggregate combination will ensure adequate adhesion.

Most of the design methods share a number of input parameters of which traffic, aggregate ALD, pavement texture and voids to be filled are the most dominant. The lack of adhesion strength as an input is attributed to the fact that there is no measurable value that relates the behaviour during the seal lifetime. This motivates the need for a test to accurately measure the bitumen aggregate bond strength and development for possible use in future design methods.

3.6 FACTORS INFLUENCING THE DESIGN & PERFORMANCE OF SEALS 3.6.1 INTRODUCTION From the different design processes above it is clear that various factors should be considered in the design of surfacing seals. This is due to the fact that these factors directly influence the behaviour of seals. The following sections include information as presented in the original text of the TRH3 (South African National Road Agency, 2007) and will aim to detail some of the most influential factors that should be considered. Figure 3-4 below illustrates the most relevant seal terms that is used throughout this study. PRE COATING

WETTING OF STONE BITUMEN BINDER BLEEDING

AVERAGE LEAST DIMENSION

VOIDS AVAILABLE SUBSTRATE

AGGREGATE

STONE EMBEDMENT

Figure 3-4 – Schematic illustration of a single seal and the terms relevant to a single seal

25 AH GREYLING

December 2012


3.6.2 SUBSTRATE Chip and spray seals are usually applied as first seals on compacted gravel or crushed stone material base material or as a reseal on varying existing surface seals. It is therefore no surprise that most important influences on the behaviour of a surfacing seal are the condition of the underlying surface and the type of material of which it is constructed. The substrate condition is one of the most important influences in determining the appropriate seal type and the type and quantity of binder to be applied as well as the size of stone to be used and the required pre-treatment.

The binder application rate will be influenced by the texture of the existing road. A new smooth pavement with low air voids will not absorb much of the binder applied to it. A dry, porous and pocked pavement surface can absorb much of the applied binder. Failure to recognize when to increase or decrease the binder application rate to account for the pavement condition can lead to excessive stone loss or bleeding.

The condition of the existing surface is generally described and measured by the visual distress discussed in Section 3.3 above as well as by the measurement of surface texture depth and unevenness, permeability and the expected embedment of the stone. The measured values can influence the seal design as follows:



Texture depth values and variations gives an indication of the additional binder required due to the surface texture and it may indicate the need for a texture treatment if the texture varies significantly.



The permeability gives an indication of the need to pre-treat the existing surfacing by adding additional binder to waterproof the porous layer.



The expected embedment of the stone gives an indication of the amount of voids in the seal that would be lost as a result of the embedment of the stone into the existing surface. In this instance, the ball penetration test provide useful information but do not take into account the softening of the existing binder by the new, sometimes very hot and oily binder applied during resealing.



The degree and extent of cracking provides an indication of the likely reflection of such cracking through the new surfacing or of the relative brittleness and loss of flexibility of the existing surfacing which need to be made up by the new seal (South African National Road Agency, 2007).

26 AH GREYLING

December 2012


3.6.3 TRAFFIC The reason behind the existence of every road is to make it possible for various forms of traffic to travel along it safely. Traffic is also the most significant influence on the performance and behaviour of seals and the number, type, speed and combination of vehicles have a significant influence on the behaviour and performance of surfacing seals.

The traffic volume, in terms of vehicles per day, plays an important role in determining the amount of bitumen binder needed to sufficiently bond and embed the aggregate. Typically, the higher the traffic volume, the lower the binder application rate will be. The number of vehicles greatly affects embedment, wearing and polishing of the stone. This in turn reduces the voids in the seal and results in flushing and a reduction in surface texture which in turn reduces the skid resistance. It is also believed that a seal requires a minimum number of vehicles per day to keep the binder elastic and flexible (South African National Road Agency, 2007).

The forces induced by heavy axle loads forces the aggregate particles to orientate towards their ALD and to lie on their flattest side. This leads to more rapid embedment of the stone into the existing surface than in the case of light vehicle loads. Embedment in turn leads to bleeding and loss a texture required for skid resistance. The applied forces and vertical stresses induced on the surfacing can be much higher than expected due to increased tyre inflation pressures.

Speed and the rate of application of forces and stresses is another major influence in the behaviour of surfacing seals. Surfacing’s under slower-moving traffic (typically less than 40km/h) generally do not perform as well as those trafficked by fast-moving vehicles. The reasons for this are the extended period of loading, horizontal stresses induced as a result of traction, particularly of heavy vehicles accelerating or breaking. Further damages include fuel and oil spillages associated with trucks and slower moving traffic.

Traffic distribution and occurrence also plays a major role in the long and short term behaviour of a seal. When traffic is concentrated in a particular wheel paths, as is typical of narrow roads and temporary deviations it will lead to accelerated embedment. High and concentrated traffic during the early life of the seal (when the binder is still soft) or during cold temperatures (when the binder is brittle) will also adversely affect the performance of the seal (South African National Road Agency, 2007). 27 AH GREYLING

December 2012


3.6.4 ROAD GEOMETRY Most design methods notes that road geometry and specifically steep gradients, sharp curves and intersection areas require specific attention during seal design and construction. Steep gradients are subjected to increased traction forces of vehicle tyres which can results in de-bonding or slippage of the surfacing. The risk of poor seal performance increases on climbs and descents because of construction difficulties at these sites. Steep gradients also cause slower vehicle speeds and that often results in flushing and bleeding.

Sharp curves cause vehicles to induce high horizontal stresses on the surfacing. These may result in ravelling and slippage of the surfacing. Vehicles on lower volume roads “ cut “ corners , resulting in the outer part of the road becoming dry and brittle, with consequent stone loss. Because of the camber on curves, higher loads are transferred to the inner side of curves, often resulting in fattiness. The situation may be aggravated by excess binder accumulated at these positions, as a result of run off during construction.

Breaking and acceleration, particularly of heavy vehicles at intersections, create high horizontal stresses in the surfacing which may result in the slippage, flushing and deformation of the surfacing. Fuel and oil spillages at intersection aggravate flushing and deformation (South African National Road Agency, 2007).

3.6.5 CLIMATE AND ENVIRONMENTAL INFLUENCES The factors affecting the seal performance through the physical environment can be summarised as follows: 3.6.5.1 CLIMATIC CONDITIONS The climatic conditions and other environmental influences is a major influence in the selection of the appropriate grade and type of binder to be used in a chip and spray seal. The following climatic influences should be considered carefully during the design of a seal: 

Hot Weather: Reduces viscosity and cohesion in binders.



Cold Weather: Some binders become hard and brittle during cold weather and this will accelerate aggregate loss and cracking.



Uncertain Weather: Large variations in temperatures require bitumen with specific properties.



UV Radiation: Accelerates the aging of binders.



Humidity: Affect the evaporation of volatiles and the breaking of emulsions , it will also accelerate ageing 28

AH GREYLING

December 2012


3.6.5.2 TEMPERATURE Bitumen is a temperature sensitive visco-elastic material. Very hot temperatures may lead to the softening of a binder and this in turn will decrease the viscosity. The decreased viscosity will lead to increased flow which will make the embedment of aggregate easier. Embedment will in turn lead to bleeding. Cold weather on the other hand can results in brittle, hard binders which will lead to subsequent aggregate loss or cracking. It is therefore very important that the correct and relevant binder be selected for each specific climatic area.

It is interesting that the Superpave Hot Mix Asphalt (HMA) Design System, used in the USA, determines both an expected high and a low design pavement temperature. The Superpave Binder specification and the supporting test procedures are products of the Strategic Highway Research Program (SHRP) completed in the USA during 1987-1992. It uses a bitumen performance grading system as penetration grading and viscosity grading are somewhat limited in their ability to fully characterize bitumen binder for use at varying temperature ranges. The Superpave performance grading (PG) is based on the idea that a bitumen binder’s properties should be related to the conditions under which it will be used. For bitumen binders, this involves expected climatic conditions as well as ageing considerations. Performance specifications will be discussed in more detail in a subsequent section (Superpave, 2012).

3.6.6 AGGREGATES Aggregates used in road building include both natural and processed gravels and rock. The following aggregate characteristics are listed in the TRH3 & TRH14 have an influence on the performance of surfacing seal: 

Shape: The shape of the aggregate influences the way it will interlock in a completed layer and this in turn influences the stability of the seal. The more angular the aggregate, the better the interlock as there is more points of contact.



Type, Size and Grading’s: The size and grading influence the voids of the matrix and in turn the amount of binder it can accommodate. A single sized tone develops good interlock which increases friction and skid resistance. With large single-sized aggregates more voids is available than in small single sized aggregates. Larger aggregates therefore allow more binder to be used, resulting in a more impermeable, longer lasting seal.



Spread Rate of Aggregate: The aggregate protects the subgrade against damage and should be applied to lie shoulder to shoulder, in a single layer and in a tightly knit pattern. If the spread rate is to low it will

29 AH GREYLING

December 2012


lead to excessive ultra-violet damage to the binder and in turn ravelling of the seal. If the spread rate is too high, the excess aggregate will be forced into the mat, leading to whip-off of the bonded aggregate. 

Average Least Dimension: The ALD is the assumed minimum thickness of the aggregate layer.



Crushing Strength: Aggregate with inadequate crushing strength will crush under the forces applied by traffic and lead to ravelling and stone loss



Resistance to Polishing: Aggregates should show resistance to polishing to ensure that friction applied by traffic does not polish the stone and lead to smoothing of the surface and in turn a loss of skid resistance.



Adhesion Properties: It is very important for the aggregate and binder to develop early and long term adhesion to ensure the aggregate is retained on the road. A loss in adhesion will lead to a loss in stone and ravelling.



Absorption /Porosity Properties: Porous aggregates absorb primarily the lighter fractions of the volatiles present in the bituminous binder. This may result in the binder becoming too brittle to retain the aggregate on the road. Porous aggregate should generally be pre coated.

Aggregate and their properties will be discussed in more detail in Chapter 6 below.

3.6.7 BITUMEN BINDERS 3.6.7.1 GENERAL The entire service life and performance of a seal is based on good adhesion between the bitumen binder, the aggregate and the road surface and on the durability and behaviour of the binder under different climatic conditions. Adhesion is one of the vital functions of the binder. Loss in retention of the aggregate, the degree of aggregate whip-off and the durability are all related to the adhesive forces developed by the binder, which depend on the type , grade and amount of binder applied (South African National Road Agency, 2007). A correctly selected binder must achieve two initial functions:

1. It must develop sufficient adhesive strength. It should be fluid enough once placed to allow for the placing and wetting of the stone and ten rapidly becomes harder to ensure retention of the aggregate. . 2. It must develop sufficient cohesive strength. The cohesive strength of the binder facilitates opening to the traffic and prevents the stone form being pulled out of or whipped off the surface.

30 AH GREYLING

December 2012


The TRH3 notes that the choice of the type of bituminous binder depends on:



Type and purpose of seal: Hot modified binder allow for a thicker application of binder. This enhances the ability to delay crack reflection and ensure rapid adhesion with aggregate when traffic has to be accommodated immediately after resealing. Modified binders are less sensitive to bleeding under high traffic volumes and have better elastic properties than unmodified bitumen. This enables modified binders to accommodate high deflections and low temperatures.



Climatic Conditions: The climatic influences were discussed previously. The TRH3 proposed binder selection included for the different climates and traffic and is shown in Table 3-2 below.



Durability of binder and long-term performance: Penetration grade bitumen, cut-back bitumen and bitumen emulsions are durable and consequently are suitable for single and double seals. Experience in South Africa has shown that polymer modified binders are superior to conventional binders.



Convenience of application: All the types of binder used in seal construction require some form of heating to bring them to a suitable spraying viscosity. Bitumen emulsions are convenient to apply, since they require less heating than other types. Some modified binders require heating up to 200°C which in turns brings special energy and safety considerations.



Compatibility with aggregate: The properties of the aggregates used for seal work in South Africa vary a great deal. A binder should be selected to ensure good adhesion with a specific aggregate to be used. For example: cationic bitumen emulsions will be found to be more suitable than anionic emulsion when aggregates such as granites or quartzite’s are used.



Traffic & Road Geometry: Please note the influence of traffic on the selection of the type of binder as specified in Table 3-2 taken from the TRH3. The influence of traffic and road geometry was discussed in the previous section and will not be discussed in more detail.

31 AH GREYLING

December 2012


Table 3-2- Recommended Binders for different Traffic and Climatic conditions (South African National Road Agency, 2007)

Once a binder is selected it should be able to withstand ‘softening-up” at higher temperatures encountered in service and to retain the stone under the action of moving wheel loads. Softening up will happen when the used binders’ softening point is lower that the operational road temperature. On the other hand, at colder temperatures, the binder should stay flexible for as long as possible to prevent reflection cracking, to accommodate road deflections and to prevent, as far as possible, the ingress of moisture into the base. The general bitumen binder properties that affect the performance of the binder on the roads can be defined as follows: 

Binder type and physical properties: The different types of binder such as penetration grade, cut back, emulsion and modified binder all behave differently under certain conditions and will all affect the performance of the seal in specific way.



Binder grade: The correct binder grade for expected climatic, pavement and traffic conditions, both under construction and long term, must be selected to ensure optimal seal performance. Each grade of binder is characterized by its own temperature/viscosity relationship. It has an optimum range over which the bitumen can be sprayed, stored, mixed or pumped. It is essential that the viscosity be kept within the range for spray application in order to obtain optimum application and it is important for a binder to have sufficient viscosity to not soften up during operational temperatures in the road.



Spray Rate: The optimal amount of bitumen is required to keep the stone on the road, (through adhesion) but to maintain enough voids to prevent bleeding. The optimum amount of binder is determined by size and shape of stone, and the volume of voids in the compacted stone layer, traffic, gradients and condition of underlying surface as per the chosen design method. 32

AH GREYLING

December 2012




Viscosity at application: Viscosity at application determines the uniformity of the application. Where bitumen is applied at too low temperature it causes streaking, due to high viscosity at the spray bar, and too high temperature it will cause binder degradation and run-off on steep gradients (South African National Road Agency, 2007).

3.6.7.2 TOWARDS PERFORMANCE SPECIFICATIONS Considering the large amount of considerations and counter consideration mentioned in the previous sections, it is no wonder that chip seal design is still considered to be some form of art. Most of the binder selection criteria has very little to do with the short term behaviour of a seal and it is mainly based on previous experiences and the elimination of certain binder types in certain conditions as problems are experienced.

This is one of the major reasons that there is currently a global movement towards bitumen binder performance specifications and the identification of binder properties that relates to actual road performances. Various studies and discussions have identified the principles underpinning the development of a performance-based specification system. These principles include: considering binder properties related to actual road performance; testing properties that reflect emulsion residues following breaking processes; evaluating properties that have quantifiable relationships to common distresses; analysing binder properties at temperatures related to actual field climates; and recognizing that distress modes differ for chip seals and other surface treatments. (Miller, 2010) Proposed specification systems currently under development in New Zealand consider the life cycle of a chip seal from a performance perspective. Table 3-3 shows the proposed performance grading system. It summarizes relevant phases, distress mechanisms and construction properties that must be controlled, proposed measured binder properties, and tested materials.

Through this system it may be possible to place minimum and maximum values on viscosities, bond-strength and binder ageing based on the performance of seals under different environmental and traffic conditions. It is in some way be related to the method described in the TRH3 as shown in Table 3-3 above. It will differ that instead of a specific binder type is will include measureable binder properties related to the physical behaviour of the seal (Opus International Consultants, 2010) .To make this a reality it should be possible to measure the specific binder properties related to the seal behaviour. It is here where the bitumen & aggregate bond strength test described and developed as part of this study may help to improve binder and aggregate selection as well as to improve the measurements of the critical binder properties during the early life phases.

33 AH GREYLING

December 2012


Table 3-3- Proposed New Zealand Performance Grading System (Opus International Consultants, 2010) Phase

Construction

Early Life 0-5 years

Mid Life 5-15 years

End of Life 15 years +

Construction Property Chip Wetting during construction

Binder Property Viscosity

Material Tested Sprayed Binder

Chip wetting and reorientation

Viscosity

Chip loss due to loss of adhesion

Bond Strength

Chip loss due to loss of cohesion

Bond Strength

Chip loss due to 'single event' cohesive failure

Bond Strength

Binder Adhesion to tires

Viscosity

Chip loss and cracking due to cohesive failure of binder

Cracking/Fatigue

Chip loss due to adhesive failure of binder

Bond Strength

Sprayed Binder/Emulsion residue Sprayed Binder/Emulsion residue Sprayed Binder/Emulsion residue Sprayed Binder/Emulsion residue Sprayed Binder/Emulsion residue +5 years aging Sprayed Binder/Emulsion residue +15 years aging Sprayed Binder/Emulsion residue +15 years aging

What is very important to note is that that no specification systems can accommodate poor construction practices. It therefore always assumed that under normal construction operations, chip seals will perform according to expected distress mechanisms. For example, specifications assume that chip seals will not be constructed during heavy rain, with excessively dirty aggregates, or after prolonged time intervals because these conditions will have deleterious effects on the final delivered product.

Miller (2010) described the four phases during the life cycle of a surfacing seal as shown in Table 3-4 above in more specific detail as part of his thesis. .He noted that during the construction phase, high binder viscosities may result in inadequate chip wetting and subsequent ravelling because binders do not sufficiently coat and wet aggregates.

Bitumen emulsions overcome this deficiency due to the water based nature of the material, so this property applies only to penetration grade binder. Following the initial construction, emulsion residue’s viscosity must remain sufficiently low to allow for aggregate reorientation under compaction and vehicle loading. Because all 34 AH GREYLING

December 2012


bitumen binders are prone to water-induced stripping, chip loss due to loss of adhesion is of primary concern during the early life of seal. Chip loss can also results from ductile or cohesive binder failures resulting from stresses and strains induced by traffic loading.

Brittle binder failures may lead to chip loss when the binder becomes too stiff during prolonged periods with cold temperatures. This is especially relevant to modified binders. Mid-life distresses occur when the road bleeds due to high temperature and traffic that reduces the binder viscosity. This may lead to the binder adhering to tires which in turn will damage the seal and expose the layers underneath. Bleeding will also result in a loss of traction and skid resistance due to the loss of surface texture due to flushing.

Durability issues related to end-of-life chip loss and cracking may result from binder fatigue, so loss of bond strength between aggregates and binders has implications for long-term durability as well. Collectively, these distresses and performance properties define the effectiveness of chip seal applications (Miller, 2010).

35 AH GREYLING

December 2012


4

BITUMEN

4.1 INTRODUCTION The South African National Standard 4001-BT1 defines bitumen as: “A non-crystalline solid of viscous mixture of complex hydrocarbons that possesses characteristic agglomerating properties, soften gradually when heated, is substantially soluble in trichloroethylene and is obtained from crude petroleum by refining processes. “

Mankind has been using bitumen in a range of applications, including mummifying the dead and as a waterproofing agent, for over 5000 years. The Shell Bitumen Handbook (2003) gives some 250 uses of bitumen. Today, the main use of bitumen is in the road construction and maintenance industry.

In a chip seal the bitumen binder are used in a combination with mineral aggregates to form waterproof layers in the various combination discussed above. As a binder, bitumen is especially valuable to the engineer because it is a strong, readily adhesive, highly waterproof and durable material. Bitumen also provides some flexibility to mixtures of mineral aggregates with which it is usually combined. The strong adhesion that occurs between the bitumen and mineral aggregates enables the bitumen to act as a binder, with the mineral aggregate providing mechanical strength for the road. Bitumen is also highly resistant to the action of most acids, alkalis and salts (Shell Bitumen, 2003).

4.2 MANUFACTURE OF BITUMEN Bitumen is manufactured during the fractional distillation of crude oil. Fractional distillation takes place in the tall steel towers usually seen at refineries. It uses the evaporation, distillation and different boiling points of all the material in crude oil to distil different fractions at various temperatures and heights within these towers.

In South Africa most bitumen used in road construction is processed at refineries in Cape Town, Durban and Sasolburg. At these refineries imported crude oils are refined to produce petrol, diesel fuel and other petroleum based products. Bitumen represents only about 2.5% of a total barrel of oil.

36 AH GREYLING

December 2012


4.3 COMPOSITION OF BITUMEN The arrangement of the internal structure of bitumen is largely determined by the chemical composition of the molecular types present. Bitumen is an intricate chemical mixture of molecules that are mainly hydrocarbons with a small amount of functional groups containing sulphur, nitrogen and oxygen atoms.

The precise composition varies according to the source of the crude oil from which the bitumen is distilled, modification and blowing during manufacture. It is therefore clear that the chemical composition of bitumen is extremely complex and a complete analysis of bitumen would be impractical. It is however possible to separate bitumen into two broader groups called asphaltenes and maltenes as indicated in Figure 4-2. The maltenes can be further sub divided into saturates, aromatics and resins.

Figure 4-1- Broad Chemical Composition of Bitumen (SABITA-M2, 2007)

4.4 TYPES AND GRADES OF BITUMEN There are six major classifications of bitumen produced by the refining and manufacturing process in South Africa.

37 AH GREYLING

December 2012


4.4.1 PENETRATION GRADE BITUMEN’S Penetration grade bitumen is the most widely used bitumen and is refined and blended to meet civil engineering and industrial specifications. The specifications are based on specific test properties that aim to account for different climatic conditions. Paving grade bitumen is also considered as the parent bitumen/base binder from which the other forms are produced. The table below indicates the South African Penetration Grade requirements (SANS4001-BT1, 2012). Table 4-1- South African Penetration Grade Bitumen Requirements

4.4.2 CUTBACK BITUMEN Cutback bitumen are in essence penetration grade bitumen blended with a small quantity of volatile solvents to reduce the viscosity for ease of handling and application, which, after the volatile solvents have evaporated, essentially reverts to the penetration-grade bitumen base (SANS4001-BT2, 2012) .Classifications include rapid curing (RC), medium curing (MC) or slow curing (SC). A cutback bitumen varies in behaviour according to the type of cutter or flux used as the diluent. White spirits is commonly used for rapid curing grades, kerosene for medium curing grades and diesel for slow curing grades.

38 AH GREYLING

December 2012


4.4.3 BLOWN BITUMEN Blown bitumen also known as Industrial bitumen or oxidised bitumen’s are made by blowing air through hot paving grade bitumen. The so-called blowing process results in harder bitumen that softens at a higher temperature than that at which paving grade bitumen softens. Blown bitumen also has more rubber-like properties and their viscosities are much less affected by changes in temperature than is the case with paving grade bitumen (BP Bitumen, 2011).

4.4.4 BITUMEN EMULSIONS Bitumen emulsion consists of bitumen particles dispersed in water in the form of an oil-in-water type emulsion. The bitumen is held in suspension by an emulsifying agent that also determines the charge of the bitumen emulsion. The two most common types are Cationic (+) bitumen emulsions that have a positive charge and Anionic (-) bitumen emulsion that have a negative charge.

The South African national Standards define bitumen emulsion as a “liquid mixture in which a substantial amount of bitumen is suspended in a finely divided condition in an aqueous medium by means of one or more suitable emulsifying agents.” (SANS 309, 2004) Due to its important role in this study, bitumen emulsion will be discussed in more detail in Chapter 5 below.

4.4.5 MODIFIED BITUMEN AND EMULSIONS 4.4.5.1 INTRODUCTION Chapter 3 has shown that the properties of standard bitumen are not always able to meet the functional and behavioural requirement when a bituminous surfacing are subjected to severe conditions such as steep gradients, very high road surface temperatures, high traffic loading or heavily trafficked inter sections, or are used on highly flexible and cracked pavements.

Under these circumstances a definite improvement in the rheological properties of the bitumen is required and this is where the different types of modified bitumen come into play. Some of the benefits that may be derived from modification include; improved consistency, reduced temperature susceptibility, improved stiffness and cohesion, improved flexibility, resilience and toughness, improved binder aggregate adhesion and improved resistance to in services ageing.

39 AH GREYLING

December 2012


4.4.5.2 MODIFICATION AND USES IN SOUTH AFRICA Modification is achieved by mixing penetration grade bitumen with polymers, crumb rubber, aliphatic synthetic wax or naturally occurring hydrocarbons. The SABITA Manual 2 classifies polymers into "elastomers" (sometimes referred to as thermoplastic elastomers) for improving the strength and elastic properties of a binder, and "plastomers" (sometimes referred to as thermoplastic polymers) for increasing the viscosity of the bitumen. However a wide range of binder properties can be improved, the most important being: 

Improved Rheology of flow characteristics



Resistance to permanent deformation and fatigue cracking due to increased elasticity



Cohesion and Adhesion



Elasticity



Reduced Ageing and Durability (SABITA-M2, 2007).

The section below is a modified extract from the original text of (SABITA-M2, 2007) and describes the types of modified bitumen that are readily available and used in South Africa: “ 

Styrene-Butadiene-Rubber (SBR) latex: SBR is a block co-polymer and can be classified as either linear or radial. It is available in the form of anionic or cationic latex, which makes blending with bitumen emulsion easier. It is also used to modify hot bitumen, but the water phase in the latex must first be re moved by boiling or foaming during the controlled addition of latex to the hot bitumen. SBR modified bitumen has been used extensively in South Africa as a cold applied bitumen emulsion in chip seals and micro surfacing, as well as a hot applied binder in chip seals and hot mix asphalt. The modified binder exhibits elastic properties which make it ideal for surfacing lightly cracked pavements.



Styrene-Butadiene-Styrene (SBS): SBS polymers are available in powder, crumb or pellet form for modifying hot bitumen. Linear as well as radial co polymers can be used depending on the end properties sought. High shear mixers are recommended for blending high polymer content binders, particularly for industrial applications. Depending on the concentration of this polymer, it increases or improves softening point, cohesive strength, elasticity, low temperature flexibility and resistance to permanent deformation. SBS modified bitumen is used in both chip seals and hot mix asphalt applications to enhance the bitumen’s all-round performance characteristics.

40 AH GREYLING

December 2012




Ethylene-Vinyl-Acetate (EVA): EVA polymers are available in pellet form and are easily dispersed in hot bitumen. EVA modified binders are thermally stable have the most pronounced effect on the binder’s end properties are its molecular weight and vinyl acetate content. EVA modified bitumen is used mainly as a plastomer in hot mix asphalt to improve rut resistance.



Bitumen Rubber: About 20% of rubber crumb (recycled tyres) is blended with bitumen at a mixing temperature of 170° - 210°C for a period of approximately one hour. During this time the aromatic oils in the bitumen are absorbed by the rubber particles, causing them to swell. It is much more viscous than unmodified bitumen, and is not a homogeneous binder, requiring special equipment for pumping and spraying. Bitumen rubber is widely used in South Africa in chip seals. On cracked and flexible pavements, bitumen rubber has resisted crack reflection remarkably well and, in spite of high application rates, its resistance to flushing has been clearly demonstrated. This resistance is due, amongst others, to its improved temperature susceptibility of viscosity. The carbon black contained in the rubber also acts as an anti-oxidant, thereby increasing the durability of the binders.



Synthetic Waxes: Long-chain hydrocarbons produced by the Fischer-Tropsch (F-T) synthesis process are used to extend the plasticity range of bitumen. Bitumen modified with F-T wax displays unique properties in that it has a lower viscosity than unmodified bitumen above 100°C, but on cooling the viscosity is higher. This enables bitumen modified with F-T wax to be sprayed and placed at lower temperatures than using conventional bitumen. The significant increase in Ring and Ball softening point of binders modified with F-T wax renders asphalt incorporating such binders to have enhanced resistance to permanent deformation and bleeding. The Figure below shows these properties in more detail:

Figure 4-2- Typical effect of F-T wax on viscosity over temperature 41 AH GREYLING

December 2012




Naturally occurring hydrocarbons: Naturally occurring hydrocarbons that are used for bitumen modification occur in deposits in North and South America, and are known as Gilsonite and Durasphalt respectively. These materials have high asphaltene contents, generally in excess of 70%, and are used to stiffen bitumen by raising the Softening Point and lowering the Penetration value. Asphalt mixes containing binders modified with these hydrocarbons display high resilient modulus values compared with those containing binders modified with polymers. The high resilient moduli achieved will enhance both the resistance to permanent deformation and the load spreading capability of the layers (SABITA-M2, 2007).”

4.4.5.3 CLASSIFICATION OF MODIFIED BINDERS The Asphalt Academy’s Technical Guideline 1 (TG1) is currently the most important authority on modified binders used in road construction. It classifies modified binders according to the four groups that include the type of application, type of modifier, type of binder system and the level of modifications. The details of these four groups can be summarised as follows: 

The type of application: Seal (S) , Asphalt (A) or Crack sealant (C)



Type of modifier: Elastomer (E) , Plastomer (P) , Rubber Crumb (R) of Hydrocarbon H)



Type of binder system: If the product is an emulsion, the letter C would follow directly after the letter indicating the type of application.



Level of modification: Usually indicates modifier content

A numerical value is used to indicate increasing softening point values. This classification system allows for a polymer-binder specification whereby the test properties for a specific class must be achieved in order to meet the specification requirements. Table 4-2 below shows the Typical Binder Classes specified in the TG1.

42 AH GREYLING

December 2012


Table 4-2- Modified Binder Classification System (Asphalt Academy-TG1, 2007)

4.4.5.4 BENEFITS OF MODIFICATIONS The primary aim of the modification of bitumen for use in surfacing seals is to increase the resistance to bleeding at high road temperatures without compromising the properties of the seal over the rest of the prevailing temperature range. (Asphalt Academy-TG1, 2007)

The use of modified bitumen is rising as a result of increases in tyre pressures, axle loads and higher traffic volumes that in turn required improved binder performance. Improved performance can be achieved in three ways:

43 AH GREYLING

December 2012




An increase in the elastic component with an associated reduction in the viscous component; and



Stiffening of the bitumen to reduce the total viscoelastic response of the layer.



Increased adhesion and cohesion.

To explain the benefits of modification it is important to understand the improvements it makes to each of the properties of normal penetration grade bitumen. The four sections below is a combination of extracts from the original text included in (SABITA-M2, 2007) & (Asphalt Academy-TG1, 2007) and it will aim give a brief overview look of the influence of modification on certain behavioural characteristics.

RHEOLOGY (FLOW BEHAVIOUR)

The TG1 defines Rheology as the study of the flow and deformation behaviour of materials. Bituminous binders behave as visco-elastic materials and their behaviour is influenced by the loading time as well as the operating temperature. At high temperatures or long loading times, binders will generally behave as viscous liquids. This will lead to bleeding and permanent deformations in asphalts. At low temperatures or short loading times, binders will behave as elastic (brittle) solids, with most deformation recovered at the end of the loading period.

The rheology of unmodified binders is relatively simple, and behaviour can be predicted through the use of simple tests such as Penetration, Softening Point and Viscosity at various temperatures. The rheology of modified binders on the other hand is highly complex, and, although the results from conventional tests may indicate a significant improvement in properties, the in-service performance of these binders is not easily categorised. (Asphalt Academy-TG1, 2007)

These properties can be explained by looking at an elastomer modified binder. The figure below shows the typical effect of an elastomer on the rheological profile of penetration grade bitumen. The following interesting observations can be made: 

At high road temperatures, say 55°C - 60°C, the modified binder has a significantly higher viscosity than, the base penetration grade bitumen and is therefore much stiffer. In conjunction with the elastic nature of the polymer network, such modified binders will exhibit a significantly higher resistance to deformation and bleeding. 44

AH GREYLING

December 2012




At the spray temperature of 130°C-180°C the modified binder also shows significantly higher viscosity. This enable a higher application of modified binder as runoff will not be a problem.



On the other had of the scale at lower service temperatures, the modified binder is more flexible than the unmodified bitumen and therefore would be less prone to cracking and becoming brittle. This can be seen from the higher penetration values at around and below 0°C. Increased flexibility and resistance to fatigue distress has also been shown to result from elastomer modification.

Figure 4-3- Effect of elastomers on the rheological profile of bitumen (SABITA-M2, 2007)

It can be deduced from the information included in Figure 4-6 above that the modified binder material would offer improved rheological performance in areas where high temperature and tensile strains are likely to occur. ELASTICITY

Elasticity can be defined as a physical property of materials which return to their original shape after the stress that caused their deformation is no longer applied. (Atanackovic & Guran, 2011) The elastic recovery of a binder is commonly used to measure the fatigue resistance of a binder or its ability to absorb large stresses without necessarily cracking or deforming.

45 AH GREYLING

December 2012


Particular modifiers have unique elastic recovery characteristics depending on the morphology of the modifier/binder system. For example, SBS as described above forms a three-dimensional network of highly elastic, butadiene chains connecting stiff styrene domains, resulting in very high elastic recoveries. SBR has random styrene/butadiene molecules, resulting in relatively lower elastic recovery.

EVA forms a rigid three-dimensional network, imparting no elastic recovery properties to the base binder but provides a high stiffness. The elastic property of a binder is generally influenced by the type of modifier as well as the degree of modification and is determined by measuring the recovery of a sample which has been extended in a low temperature ductilometer. In general, there is a direct relationship between elastic recovery and the degree of modification”. (Asphalt Academy-TG1, 2007) COHESION

The TG1 defines cohesion as a measure of the tensile stress required to break the bond between molecules of the bituminous binder. The inherent strength, tenacity and toughness of the bituminous binders are improved by modification with thermoplastic polymers and rubber crumbs. Hence, a greater force or tensile stress is required to break the molecular bonds of modified binders and cause failure compared with a lower tensile stress required to break the bonds of conventional binders.

A force-ductility test is used to determine the cohesive strength of a modified binder and involves the elongation of a sample with the force measured at very small elongation intervals. Figure 4-7 shows a graph of the typical profile of various modified binder types obtained during the test.

As shown in Figure 4-7, the maximum force is reached early in the elongation process. The elastic phase is represented by the area before the initial peak and the total area under the curve can be used to calculate toughness. This is a good indication of the energy required to extend the binder and therefore provides a good estimation of resistance to cracking.

46 AH GREYLING

December 2012


Figure 4-4- Typical force-ductility curves for various modified binders (Asphalt Academy-TG1, 2007)

The energy required to elongate elastomeric modified binders is generally significantly more than that for conventional binder. Plastomeric modified binders will impart stiffness to the bituminous binders but not necessarily improve its cohesive nature. Such modified binders may well perform in a brittle manner in tension. The cohesive properties of modified binders provides direction to design engineers on how soon after construction a seal could be opened to traffic as well as providing an assessment of the ability of the binder to withstand shear stresses imparted by heavy traffic. (Asphalt Academy-TG1, 2007) ADHESION

The TG1 further describes and define adhesion as the measure of the stresses required to break the bonds between the bituminous binder and mineral aggregate. Adhesion is largely dependent on the physical chemistry as well as the chemical nature of the bituminous binder and aggregate type when combined for application. The following factors have an impact on adhesion at the stone/bitumen interface: 

The presence of dust and/or moisture which could reduce adhesion at the bitumen/aggregate interface.



The level of modification that influences the viscosity of bitumen which in turn affects the wetting ability or time to coat the road stone with bituminous binder.



Ambient road and air temperatures and, especially overnight temperatures. 47

AH GREYLING

December 2012


Generally by increasing the film thickness of the binder layer the adhesion can also be improved. Temporary reduction of the viscosity by the addition of cutters during colder weather conditions also improves the adhesion properties by keeping the binders flexible. Care should however be taken in areas with hot humid climates and/or heavy traffic conditions as binders with cutter may tend to bleed.

It will be discussed in more detail in Chapter 6 but it should also be noted that different types of aggregates will exhibit different adhesion behaviour depending on the chemical nature of the parent rock in terms of its hydrophilic (water-attractive) or oleo phobic (oil-repelling) nature. Bitumen is by nature oleophilic (oil-attracting) or hydrophobic (water-repelling).

Therefore, based on the inherent character of the aggregate it may or may not react (form chemical, chargerelated bonds) with water but the presence of water will have a negative influence on the adhesion properties and it will repel the bitumen. Acidic aggregates are more hydrophilic than basic aggregates. Acidic aggregates will therefore have poor adhesion properties in the presence of water. Cationic spray grade emulsion overcomes this tendency when the free electrons on the aggregate form physical/electrical bonds with the positively charged bitumen and SBR latex droplets as shown if Figure 4-8. (Asphalt Academy-TG1, 2007)

Figure 4-5- Adhesion characteristics of Cationic latex modified emulsions (Asphalt Academy-TG1, 2007)

48 AH GREYLING

December 2012


5

BITUMEN EMULSIONS

5.1 BACKGROUND In a paper by Kobus Louw published in 2004 he described the origin of bitumen emulsion as follows: “In 1922, two British scientists patented their invention of bitumen emulsion, which was further improved the following year when a new patent was filed in France. The use of bitumen emulsions rapidly spread throughout the world and by 1928 the first emulsion factory in South Africa, was erected in Bellville by the company Colas. The first emulsions manufactured were all anionic in character and prepared from the soaps formed by neutralising long chain fatty acids such as tall oil, oleic acid, stearic acid and napthenic acids. Due to the neutralisation of the fatty acids by sodium or potassium hydroxide, anionic emulsions are alkaline in character and impart a negative charge to the dispersed bitumen droplets. It was long realised that the use of anionic emulsions presented adhesion problems with certain difficult aggregates such as quartzite’s and granites. The development of cationic emulsifiers, based on long chain amines, largely overcame these problems during the late 1950’s. The long chain amines are neutralised with mineral acids such as hydrochloric, sulphuric and sometimes phosphoric acid, and the resultant acidic emulsifier imparts a positive character to the bitumen emulsion (Louw, et al., 2004).”

5.2 DESCRIPTION Bitumen is difficult to work with at ambient temperatures since it is a highly viscous material under these conditions. It can, however, be transformed into a workable state by either applying heat (hot mixes), by blending with petroleum solvents (cutback mixes) or by emulsification with a surfactant in water to form bitumen emulsion (Gorman, et al., 2004).

An emulsion is a dispersion of small droplets of one liquid in another liquid. Typical examples include such everyday products as milk, butter, mayonnaise, and cosmetic creams. Emulsions can be formed by any two immiscible liquids, but in most emulsions one of the phases is water.

Oil-in-water emulsions are those in which the continuous phase is water and the dispersed (droplet) phase is an “oily” liquid. Water-in-oil “inverted” emulsions are those in which the continuous phase is oil and the disperse phase is water. Emulsions can also have more complex structures.

49 AH GREYLING

December 2012


Bitumen emulsions are therefore dispersions of bitumen in water. It is manufactured by processing hot bitumen, water and emulsifier in a high speed colloid mill that disperses the bitumen in the water in the form of small droplets. These droplets or particles of bitumen are normally in the 5-10 micrometre size range but may be even smaller.

The emulsifier assists in forming and maintaining the dispersion of fine droplets of bitumen. Bitumen emulsions normally comprise between 30% and 80% bitumen by volume and have a lower viscosity compared to the bitumen from which they are produced. This makes emulsion workable at ambient temperatures.

Emulsions are brown liquids with consistencies from that of milk to double cream, which depend mostly on the bitumen content and the particle size (Gorman, et al., 2004). The figure below shows a micrograph of a bitumen emulsion:

Figure 5-1- Micrograph of Bitumen Emulsion (James, 2006)

Bitumen emulsions are generally available in either Cationic or Anionic Emulsions. The terms cationic and anionic is derived from the electrical charges on the bitumen globules. In an anionic emulsion the bitumen particles are negatively charged. In a cationic emulsion the bitumen particles are positively charged. Cationic emulsions are more widely used as they have superior adhesive properties to a range of mineral aggregates (Gransburg, et al., 2010).

50 AH GREYLING

December 2012


5.3 MANUFACTURING OF EMULSIONS Emulsions are manufactured by mixing hot bitumen with water containing emulsifying agents and applying mechanical energy sufficient to break up the bitumen into droplets. The mechanical energy is applied using a colloidal mill as shown in Figure 5-2 below. A colloidal mill consists of a high speed rotor revolving at 10006000 rpm in a stator. The clearance between the rotor and stator can usually be adjusted between 0.25 and 0.5 μm.

Figure 5-2- Colloidal Mill (Jaixing Mide Machinery Company, 2012)

The emulsion manufacturing process is shown in a schematic from the TG2 below:

Figure 5-3- Bitumen Emulsion Manufacturing Process (Asphalt Academy-TG2, 2009) 51 AH GREYLING

December 2012


The water phase is prepared whereby the emulsifier chemicals are dissolved in heated water and either mixed (soaped) with alkalis in the case of an ionic emulsion, or acidified with inorganic acids for cationic emulsions. The temperature of the bitumen phase should be strictly controlled to ensure that its viscosity is within the appropriate range for emulsification, and to prevent boiling of the emulsion on exit from the colloid mill.

The base bitumen normally used is 70/100 penetration grade. The stability of the emulsion is determined primarily by the type and quantity of the emulsifying agent used. Mixing any anionic emulsion with any cationic emulsion in any proportion will cause the mixture to "break", i.e. separate into water and bitumen, almost immediately.

5.4 TYPES AND CLASSIFICATION OF EMULSIONS Bitumen emulsions are classified according to the sign of the charge on the emulsified bitumen droplets and according to their reactivity and rate of setting. Cationic emulsions have droplets which carry a positive charge (+). Anionic emulsions have negatively charged droplets (-).

Rapid-setting (RS) emulsions set quickly in contact with clean aggregates of low-surface area, such as the aggregates used in chip seals. Medium-setting (MS) emulsions set sufficiently less quickly that they can be mixed with aggregates of low surface area, such as those used in open-graded asphalt mixes. Slow Setting (SS) emulsions will mix with reactive aggregates of high surface area.

In the USA emulsions are named according to the ASTM D2397 - 05, 2012 and ASTM D977-12, 2012 standards. Cationic Rapid Set, Cationic Medium Set and Cationic Slow Set emulsions are denoted by the codes CRS, CMS, and CSS, whereas anionic emulsions are just called Rapid Set , Medium Set , and Slow Set, followed by numbers and text indicating the emulsion viscosity and residue properties.

In South Africa the SANS 309, 2004 and SANS 548, 2003 specifications are relevant. These standard divide both anionic and cationic into three usage specific categories as defined below: 

Pre-mix type emulsions is emulsion with sufficient stability to allow mixing with certain types of aggregate before breaking of the emulsion occurs. (Medium Set)



Spray type emulsions is characterized by rapid breaking of the emulsion on application, and is normally unsuitable for mixing with crushed aggregate. ( Rapid Set) 52

AH GREYLING

December 2012




Stable-mix type emulsions have sufficient mechanical and chemical stability for all purposes involving mixing with stone chippings, natural gravels, and soil. This includes aggregates containing large proportions of fines or chemically active materials such as cement or hydrated lime. (Slow Set)

The tables below indicate the specification requirements of Anionic and Cationic Emulsions. Table 5-1- South African Anionic Bitumen Emulsion Specification (SANS 309, 2004)

Table 5-2- South African Cationic Bitumen Emulsion Specification (SANS 548, 2003)

53 AH GREYLING

December 2012


5.5 BITUMEN EMULSION PROPERTIES The emulsification of bitumen into bitumen emulsions may decrease the viscosity but the addition of emulsifiers gives emulsion certain specific properties that control their behaviour. The section below will have a look at the chemical properties and their influence on the breaking mechanism of emulsions.

5.5.1 CHEMICAL PROPERTIES The emulsifiers used in bitumen emulsion can be classified as surface active agents or surfactants. Surfactants have non-polar lipophilic (oil-loving) and polar hydrophilic (water-loving) portions in the same molecule. These molecules concentrate at the interface between the water and bitumen and are orientated with the polar group in the water and the nonpolar parts of the molecule in the oil. The figure below shows a visual representation of these molecules. (James, 2006)

Figure 5-4- Cationic Emulsifier Molecule (James, 2006)

The orientation of the molecules reduces the energy required to emulsify the bitumen and prevents coalescence of the droplets once formed. The choice and concentration of emulsifier also largely determines the charge on the bitumen droplet and the reactivity of the emulsion produced (Boussad & Martin, 1996). From Figure 5-4 above it can be seen that a typical emulsifier has a hydrophilic “head” group and lipophilic (hydrophobic) hydrocarbon “tail” comprising 12 to 18 carbon atoms. This hydrocarbon tail is represented by “R” in chemical formulas. It is most often derived from natural fats and oils, tall oil, wood resins, or lignin.

54 AH GREYLING

December 2012


Emulsifiers can be classified into anionic, cationic, and non-ionic types depending on the charge their head groups adopt in water, although this charge may also depend on pH (James, 2006) .Table 5-3 below details the chemistry of bitumen emulsifiers. Table 5-3- Chemistry of Bitumen Emulsifiers (James, 2002) Lipophilic Portion (Hydrophobic) Tallowalkyl-

Head Group

Counter Ion

[-NH2CH2CH2CH2NH3]2+ +

Tallowalkyl-

[-N(CH3)3]

Nonylphenyl-

[-O(CH2CH2O)100H]

Tall Oil -

[-COO]

Alkylbenzene

[-SO3]-

2 Cl-

Cl

None

-

Na

+

Na+

Head Group Charge at pH=2

Head Group Charge at pH=11

(Acidic)

(Alkaline)

Positive

Neutral

Positive

Positive

Neutral

Non Ionic

Neutral

Negative

Negative

Negative

Cationic emulsifiers are ammonium compounds that contain positively charged nitrogen (N) atoms in their head group; anionic emulsifiers typically contain negatively charged oxygen (O) atoms. The two Figures below shows the typical chemical formulas of Anionic and Cationic emulsions. Note the head and tail groups as described above:

Figure 5-5- Anionic Emulsifier (Bickford, 2001)

55 AH GREYLING

December 2012


Figure 5-6- Cationic Emulsifier (Bickford, 2001)

Several studies have shown that even non-ionic emulsifiers produce emulsions whose droplets have a small negative charge in water and non-ionic emulsifiers are often used in slow-setting bitumen emulsions. In the case of bitumen emulsions, the bitumen itself contains surface active species which can also concentrate at the interface. The size and sign of the charge on the bitumen droplets can be measured and is expressed as the “zeta potential” of the droplet (Wates, 1993).

The zeta potential is strongly pH-dependent both because of the pH dependence of the charge on the emulsifier and also because polar components of the bitumen itself may ionize. Zeta potential measurements show that the charge on the bitumen droplets becomes more negative as the pH rises. As the concentration of the emulsifier increases, the particle size of the emulsion is reduced. Slow Setting emulsions, which contain higher concentrations of emulsifier, generally have smaller particle size than Rapid Setting grades (Rabiot & B. G. Koenders, 1999).

Emulsifiers are often supplied in a water-insoluble form to the emulsion producer and need to be neutralized with acid or alkali by the emulsion manufacturer to generate the anionic or cationic water-soluble form used to prepare the soap solution. The choice of the acid or alkali and the final pH of the emulsion influence the emulsion properties. Hydrochloric acid and occasionally phosphoric acid are the acids used, and sodium and potassium hydroxide are the most common alkalis. Cationic emulsions are usually acid, and anionic emulsions are typically alkaline. Table 5-4 below shows some typical emulsion recipes.

56 AH GREYLING

December 2012


Table 5-4- Typical Emulsion Recipes (James, 2006) Cationic Rapid Setting

Cationic Slow Setting

Anionic Rapid Setting

Anionic Slow Setting

(CRS65)

(CSS60)

(RS65)

(SS60)

Bitumen

65

Tallowdiamine

0.2

Bitumen Tallow diquaternary ammonium

Hydrochloric

0.15

acid, 35%

60

Bitumen

65

Tall Oil

0.3

Bitumen Ethoxylated nonyl phenol

60 0.5

0.6 Sodium

chloride

hydroxide

0.2

Lignins

0.5

Soap pH

1.5-2.5

Soap pH

3-7

Soap pH

11-12

Soap pH

10-12

Water

To 100

Water

To 100

Water

To 100

Water

To 100

Increasing the emulsifier concentration decreases the reactivity of the emulsion. Medium Set emulsions are generally formulated with the same emulsifiers as Rapid Set grades but at higher concentration (0.4%–0.8%). The emulsion producer can adapt the emulsion recipe to cope with reactive aggregates or high temperatures, generally by increasing the emulsifier concentration or blending emulsifiers of lower reactivity (James, 2006).

5.5.2 BREAKING OF EMULSION Emulsified bitumen must revert to a continuous bitumen film in order to act as a binder in a surfacing seal or other application. This typically involves flocculation and coalescence of the bitumen droplets and removal of the water from the applied emulsion. There has been a lot of research on the specific mechanisms that is responsible for the breaking of bitumen emulsions (James, 2002) .Consensus has been reached that breaking takes place due to one or more of the following mechanisms:



Adsorption of free emulsifier on aggregate.



Rise in pH caused by aggregate or cement.



Loss of water.



Adsorption of bitumen particles on aggregate.

The figure below is a visual representation of the possible stages of setting of a cationic emulsion

57 AH GREYLING

December 2012


Figure 5-7- Possible stages in the setting of cationic emulsion (James, 2002)

The process above is also described in the Transport Research Circular Number E-C102 (James, 2006) as follows: “ 

Free emulsifier adsorbs onto the (oppositely charged) mineral surface, which neutralizes some charge on the surface while at the same time making the surface somewhat lipophilic. Too high a free emulsifier concentration in relation to the surface area of the aggregate can actually reverse the charge on the minerals and so inhibit the setting of the emulsion.



Minerals neutralize acids in the emulsion, causing loss of charge on the emulsion droplets, leading first to flocculation of the asphalt droplets and then to a slower coalescence of the droplets.



The coalescence of the droplets results in a continuous film of binder, with some minute water droplets initially being trapped in the coalesced layer. Water is adsorbed by the mineral, as well as evaporates from the system.



Droplets in contact with the mineral spread on the surface, especially that surface made lipophilic by adsorbed emulsifier, eventually displacing the water film on the aggregate surface.”

The speed of the breaking, setting and curing processes depends on the reactivity of the emulsion, the reactivity of the aggregate and environmental factors, such as temperature, humidity, wind speed, and mechanical action. The setting or breaking of a bitumen emulsion can therefore be divided into four distinct phases as per James (2006) and are depicted in Figure 5-8 on the next page.

58 AH GREYLING

December 2012


Figure 5-8- Typical stages in the breaking of emulsions (James, 2006)



In the first phase the emulsion is stable and the distance between the dispersed particles is sufficient to prevent contact or possible coalescence.



In the second phase there is a close approach between the droplets which can be caused by destabilisation of the emulsifier phase, evaporation of the water phase or mechanical action. This phase is called the flocculation phase.



In the third phase the close proximity of the bitumen droplets, cause them to fuse and flow together.



The final phase is the coalescence phase where the last of the water is fused out and the bitumen residue is in its final form.

Each of these phases is influenced by different chemical and environmental factors. Important factors in the breaking process are as follows:



Changes in pH caused by reaction of the aggregate with acids in the emulsion



Adsorption of free emulsifier onto the aggregate surface



The flocculation of the emulsion droplets with the fines.



Evaporation of water increases emulsifier content which leads to flocculation



Mechanical action, such as compaction or traffic, may squeeze the droplets together, promoting coalescence and squeezing water out of the coalesced film (Walter, 2002).

59 AH GREYLING

December 2012


5.5.3 ADHESION Aggregates take up a characteristic surface charge in water which depends on the nature of the minerals, the pH, and the presence of soluble salts. So-called “acid” aggregates high in silica tend to take up a negative charge. The Figure below is a simplified visual representation of the reaction of Cationic and Anionic Emulsion on “Acid” aggregates. From this is can be seen that Cationic emulsion will form a strong tight bond with “Acid” aggregates as the positive charges of the emulsion will be strongly attracted by the negative charge of the “Acid” aggregate.

Figure 5-9- Emulsion reaction with Acid Rock Types (Bickford, 2001) Other, none “Acid” aggregates like carbonates, and fillers, like cement, may neutralize the acid in cationic emulsions causing the pH to rise and the emulsion to be destabilized. The figure below is a simplified visual representation of the reaction of Cationic and Anionic Emulsion on “Basic” aggregates.

Figure 5-10- Emulsion reaction with Acid Rock Types (Bickford, 2001) From the figures above it can be concluded that Cationic Emulsion will form stronger bonds with Acidic rocks than Anionic Emulsion would. This will be tested and confirmed in subsequent Chapters of this report.

60 AH GREYLING

December 2012


6

AGGREGATES

6.1.1 INTRODUCTION The section above briefly touched on acid and basic rocks and it is clear that there is a significant influence on bitumen and bitumen emulsions behaviour when they come into contact with different aggregates. The term aggregate generally refers to crushed rock but other materials may also be used. Aggregates are the second major component of any chip and spray seals as it bonds with the bitumen binder to form a strong, durable and waterproof surface that must fulfil the following six major functions:



The texture, angularity of a chipped matrix of aggregate has a specific texture that provides a skid resistance surface.



Rock aggregate has a natural resistance to abrasion and polishing that is expected on a road due to moving wheel loads and traction forces.



The resistance to crushing also makes the transfers the wheel loads to the underlying pavement structures possible.



The chipped aggregate matrix also provides a strong skeletal structure to accommodate the elastic and impermeable bituminous binder.



This skeleton structure needs to have sufficient strength and voids to prevent the binder from flushing to the surface under loading.



Aggregate also protect the bitumen binder from the harmful ultra-violet rays of the sun (Milne, 2004).

6.1.2 TYPES OF AGGREGATE Rock and crushed rock aggregate types used in road construction have historically been classified as either “Acid” or “Basic” rocks. The geology and chemistry of rocks is unfortunately not that simple. Rocks are generally classified by mineral and chemical composition, by the texture of the constituent particles and by the processes that formed them. These indicators separate rocks into three major rock types: igneous, sedimentary, and metamorphic.

Igneous rocks are formed when molten magma cools and are divided into two main categories: plutonic rock and volcanic.

61 AH GREYLING

December 2012


Plutonic rock or intrusive rocks result when magma cools and crystallizes slowly within the Earth's crust (e.g. granite), while volcanic or extrusive rocks result from magma reaching the surface either as lava or fragmental ejected rock (e.g. pumice and basalt)

Sedimentary rocks are formed by deposition of clastic sediments, organic matter, or chemical precipitates (evaporates), followed by compaction of the particulate matter and cementation during diagenesis. Sedimentary rocks form at or near the Earth's surface. Typical distribution and examples include:



Mud rocks comprise 65% (mudstone, shale and siltstone)



Sandstones 20 to 25%



And Carbonate rocks 10 to 15% (limestone and dolerite)

Metamorphic rocks are formed by subjecting any rock type (including previously formed metamorphic rock) to different temperature and pressure conditions than those in which the original rock was formed. These temperatures and pressures are always higher than those at the Earth's surface and must be sufficiently high so as to change the original minerals into other mineral types or else into other forms of the same minerals (e.g. by recrystallization) (Blatt & Tracy, 1996) . Examples of Metamorphic rock include Quartzite and Hornfels.

The table below was extracted from “The natural road construction materials of South Africa” (Weinert, 1980) and it shows the characteristic combinations of rock forming materials in rocks and well as examples of typical rocks in each of the rock group.

62 AH GREYLING

December 2012


Table 6-1- Characteristic combinations of rock forming materials in rocks (Weinert, 1980) Characteristic combinations of rock forming materials in rocks Major Rock Group

Igneous

Sub Group

Characteristic Minerals

Typical Rock

Acid

Quartz, Orthoclase, Mica or amphibole

Granite

Orthoclase, amphibole

Syenite

Plagioclase, Amphibole

Diorite

Basic

Plagioclase, Amphibole

Norite,Dolerite

Ultra-Basic

Pyroxene,Oliviae

Pyroxenite,Peridotite

Quartz

Sandstone

Quartz, Orthoclase

Arkose

Quartz and incidental others

Conglomerate

Clay mineral , some quartz

Calcite

Shale Tillite, Greywacke, volcanic ejecta Limestone

Dolomite

Dolomite

Opal and/or chalcedony

No Minerals

Chert Gypsum and other salt deposits Coal, oil

Quartz, Orthoclase ( occasionally mica)

Gneiss

Intermediate

Clastic

Quartz, clay mineral and incidental others Sedimentary Chemical Precipices

Various Salts Organic

Metamorphic

Subdivisions are complicated and of little relevance to the engineering properties of rock

Quartz, Muscovite ( occasionally biotite)

Mica Schist

Amphibole

Amphibolite

Quartz

Quartzite

Calcite Amorphous silica or quartz and various others

Marble Hornfels

The classification of aggregate as acid and basic in the road construction industry was originally done in the TRH7 published in 1972. Weinert explained that the basis for the TRH7 classification was not always correct due to the fact that original classification was based on the assumption by earlier petrologists the silica (SiO2) is the radical of silicic acid (Weinert, 1980).

Weinert proposed the chemical concept that a rock (or aggregate) is acid if the acidic compounds, SiO2 and CO2 comprise more than 60 percentage mass as determined in a chemical analysis. The table below shows this classification for some of the better known and used aggregates

63 AH GREYLING

December 2012


Table 6-2- Rock Classification System (Weinert, 1980) ROCK NAME

ALTERNATIVE NAME (OLD TRH7)

% BY MASS OF SUM OF SiO2+CO2

Quartz

Quartzite

100

Quartzitic sandstone

Sandstone

94.6

Vein quartz

Quartzite

64.26

Sparagmite

Tillite

80.9

Arkose

Sandstone

75.9

Granite

Granite

70.18

Greywacke

Tillite

65.8

Shale

Hornfels

60.7

Syenite

Syenite

60.19

Andesite

Andesite

59.59

Quartz basalt

Basalt

55.46

Quartz

Norite

54.39

Magnesite

Dolomite

52.38

Trachy-dolerite

Dolerite

49.2

Basalt

Basalt

49.06

Gabbro

Norite

48.27

Dolomite

Dolomite

47.82

Limestone Marble

Marble

46.77

Basanite Basalt

Basalt

44.64

Calcite Marble

Marble

44

Nephelinitc Dolerite

Dolerite

41.17

CLASSIFICATION

ACIDIC

SLIGHTLY ACIDIC

BASIC

The TRH14 published in 1985, also states that bitumen as a rule does not adhere well to the “acidic” rocks referred to above. Weinert concluded that the strength of adhesion of bitumen to aggregate depends, as far as the stone is concerned, on its surface texture and not due to its Acidity. He reasoned that the more unevenly textured the surface, the stronger the adhesion is, and the smoother the surface the weaker the adhesion (Weinert, 1980). It is assumed that he did not include Bitumen Emulsions in this theory as it has been shown that the acidity of aggregate influences the breaking of emulsion types (Louw, et al., 2004).

As noted above, the quantity of silica (SiO2) or carbon dioxide (CO3) and whether the rock is 'acid' or 'basic', is of little importance. All rocks are weakly negatively charged and there is no noticeable difference in regard to the total charge between the different types of rock. Weinert goes on to make the following very interesting statement.

64 AH GREYLING

December 2012


“Considering the number of charged i.e. ionized, atoms in a rock, just more than 30 per cent of the atoms in a rock are anions, all being oxygen ions and just more than 20 per cent are cations. These values are very much the same for all types of rock. In general, therefore, all rocks have a weak negative charge and there is no justification for explaining the different adhesive properties of binders through a difference between 'acid' and 'basic' rocks.”

The section below is based on extracts from the original text of Weinert book (Weinert, 1980).

“The chemical composition of rocks has, however, an indirect influence on the surface texture which is obtained after crushing. If only a limited number of chemical components make up the bulk of the mass of a rock, the number of different types of mineral in such a rock is small and the individual minerals may reach a relatively large size. In most cases, silica (SiO2) is the most prevalent material component in rocks. In this case the properties of the rock may be superior to such a degree that the other component has little or no influence.

The resulting rock is of that type which is commonly called 'acid', e.g. granite but also quartzite, and which is normally characterized by rather large minerals, particularly the orthoclase feldspar in the case of granite. On crushing, such a rock produces many smooth faces or smooth areas within such faces. If silica becomes so pronounced that it is the only or nearly the only component of the rock, as in vein quartz, the extremely strong bond between the quartz minerals causes a fracture, as developed during crushing, to pass right through the minerals and smooth surfaces are again the result. If much or most of the rock consists of amorphous silica and its bond with other minerals, particularly quartz, is sufficiently strong, e.g. in quartzite or hornfels, smooth crushing faces are again produced.

Also in cases where another component is present in above average quantity, e.g. potassium in syenite, large individual minerals (orthoclase feldspar in the case of syenite) develop and crushing leads to smooth, flat faces. There is no such tendency to produce flat, smooth crushing faces in ‘basic’ rocks which contain large sized minerals only in exceptional cases. These differences in the type of crushing face formed in the different types of rock have probably led to the impression that ‘acid’ rocks cause more problems with adhesion than do ‘basic’ ones.”

65 AH GREYLING

December 2012


Road building experience has shown, no matter what the reasons, that that not all rock types adhere equally well to bitumen and there is definite difference in the affinity of acidic and alkaline rocks to bitumen and bitumen emulsions. This can in part be explained by the interaction of emulsion and aggregates as discussed in the previous section. This is also confirmed by the following extract from a paper by Kobus Louw published in 2004.

“Basic aggregates such as limestones are slightly soluble in water, raising the pH of the water phase which will destabilise the emulsifier system of a cationic emulsion, resulting in the initiation of coagulation. During the dissociating of limestone, calcium ions are also released into the water phase. When anionic emulsions are used with limestone aggregates, the calcium ions can destabilise the anionic emulsifier system, promoting the on-set of coagulation. (Louw, et al., 2004)”

6.1.3 GENERAL Based on the information above it is clear that aggregates crushed from different rocks will show varying behaviour in combination with different binders. These influences include the mineralogy of the rocks, whether it is “Acid” or “Basic” and the final surface texture of the crushed aggregates. These influences will be discussed and evaluated in Chapter 9 below.

It must be noted that various types of both acid and basic aggregates are used in South Africa to construct seals. The major influence on the type of aggregate used is usually the aggregate availability. Aggregate sources are generally situated near large cities and town centres or in strategic positions in rural areas. Due to the vast distances of the South African road network, coupled with the cost involved of hauling material over long distances, the closest available source is usually used. It is almost always easier and more economical to change the binder than it is to change the aggregate sources.

During the testing phase of this project the properties of granite and tillite will be investigated in more detail.

66 AH GREYLING

December 2012


7

TEST METHODS AND EQUIPMENT

7.1 INTRODUCTION The information discussed in the preceding section showed how important it is in the early life of a chip seal for the bitumen binder and aggregate to develop adequate adhesion. This is critical to ensure that no premature failure takes place due to adhesion-related problems. The need for a laboratory test capable of measuring the required information is therefore critical. As part of his thesis Mr Timothy Miller described the need as follows:

“A number of different types of test have been developed to compare combination of bitumen, aggregate and water. While some tests measure performance properties, it is clear from the literature review that existing test methods do not capture the full spectrum of chip seal performance over the entire chip seal life cycle. Common tests related to emulsion composition; consistency and material stability assess binder production but do not address field performance. Therefore, the focus of this investigation is on the test methods and equipment needed to evaluate performance properties, as this is seen as the greatest obstacle to instituting performance-based specification systems” (Miller, 2010)

The section below will evaluate the current adhesion tests available as well as comment on their suitability to test the bitumen-aggregate adhesion bond strength development.

7.2 ADHESION TESTS The Shell Bitumen Handbook places adhesion testing into a number of categories. There may be several different tests within each type but, in most cases, the individual tests of one type differ in detail rather than in principle (Shell Bitumen, 2003) .These categories can be summarised as follows: 

Static immersion & Dynamic immersion tests



Chemical immersion tests



Immersion Mechanical & Immersion Trafficking tests



Coating Tests



Absorption tests



Impact tests



Pull –off Tests

67 AH GREYLING

December 2012


7.2.1 STATIC IMMERSION TESTS This is the simplest type of test and consists of aggregate being coated with bitumen that is then immersed in water. The fundamental problem with this method is its subjective nature and the fact that it is completely operator dependant and has poor repeatability. It must also be noted that in some cases, an aggregate with good laboratory performance may perform poorly on some occasions and those with poor static immersion test results may perform satisfactory in practise. An example of a static immersion test is the British Standard EN 13614:2011, Bitumen and bituminous binders. Determination of adhesively of bituminous emulsions by water immersion test.

7.2.2 DYNAMIC IMMERSION TESTS This type of test is very similar to the static immersion test with the only difference being that the sample is agitated mechanically by shaking or kneading. The degree of stripping is again estimated visually and repeatability is very poor and results are operator dependant. An example of a dynamic immersion test is the British Standard EN 12274-7:2005 Slurry surfacing. Test methods. Shaking abrasion test.

7.2.3 CHEMICAL IMMERSION TESTS In this type of test, aggregate coated in bitumen is boiled in solutions containing various concentration of sodium carbonate. The strength of sodium carbonate solution in which stripping is first observed is used as a measure of adhesion. However the artificial condition of the test is unlikely to predict the likely performance on the road. An example of a chemical immersion test is the TMH 1 - Method B11 -The determination of Adhesion of Bituminous Binder to Stone Aggregate by means of the Chemical Immersion test also known as the Riedel & Weber test.

7.2.4 MECHANICAL AND IMMERSION TRAFFICKING TESTS Immersion mechanical tests and Immersion trafficking tests is mostly relevant to Asphalt and include test like the Retained Marshall Stability Test, the Retained stiffness test, retained Cantabro test and the Immersion wheel tracking test. These tests are not relevant for chip seals and will therefore not be discussed in more detail.

68 AH GREYLING

December 2012


7.2.5 COATING TESTS This type of testing aims to assess the adhesion between aggregate and bitumen in the presence of water. In the Immersion Tray Test from the British Road Note 39 (Roberts & Nicolls, 2008), aggregate chips are applied to a tray of bitumen covered with a thin layer of water. By careful examination of the chippings, it may be possible to determine whether surface active agents improve adhesion under wet conditions.

7.2.6 ADSORPTION TESTS As part of the Strategic Highways Research Program (SHRP) in the USA, the Net Absorption Test method was developed to evaluate moisture damage in asphalts (Curtis, et al., 1993) .This test is an extremely complicated test and it requires a ( not so readily available) spectrophotometer to perform measurement. It is not a commonly used test in South Africa and will not be discussed in further detail.

7.2.7 IMPACT TEST The Vialit pendulum Test and the Vialit Plate Test is generally the only two test available to measure bitumen adhesion with impact tests. Both these methods are easily adapted to measure a wide range of conditions. The test are relevant in situations where aggregate is in direct contact with traffic stresses and it is therefore ideal for test related to chip seals.

The Vialit Pendulum Test is shown in the picture below. The procedure involves placing a thin film of binder between two cubes and measuring the energy required to remove the upper block.

Figure 7-1- Vialit Pendulum test from the Shell Bitumen Handbook (Shell Bitumen, 2003)

69 AH GREYLING

December 2012


For the Vialit Plate Test, aggregate particles are pressed onto a steel plate of bitumen. The plate is the turned upside down and a steel ball is dropped onto the reserve side. The impact of the ball may cause detachment of the aggregate particles depending on the test conditions. The number of detached aggregate chips vs. the number of impacts can be used as an indication of performance. Visual inspection of the detached aggregate can usually determine the type of failure. In South Africa a modified version of the Vialit Plate test is used as per Method MB-7 of the TG1 (Asphalt Academy-TG1, 2007)

7.2.8 PULL -OFF TESTS The Shell Bitumen Handbook list two pull-off tests used to asses bitumen adhesion. The first is the Instron pull-off test which uses an Instron Tensile apparatus to extract aggregate test specimens from containers of bitumen under controlled laboratory conditions. The second is the limpet pull-off test that was developed to measure, the bond strength between the aggregate of a chip seal and the base course. The test consists of a 50mm diameter steel plate that is fixed to the road surface and the maximum load to achieve pull off is achieved.

The forum of European National Highway Research Laboratories in the BiTVal Phase 1 Report published in 2004 (Forum of European National Highway Research Laboratories, 2004) notes the pneumatic adhesion tensile testing instrument (PATTI) as a possible adhesion tester. This method was originally developed to test the pull-off strength of a coating on rigid substrates such as metal, concrete or wood as per ASTM D 4541-02 and was used as early as the middle 1990’s to evaluate the adhesive loss of bitumen binder-aggregate systems exposed to water (Youtcheff & Aurilio, 1999).

The sketch on the next page shows a schematic cross section of the original PATTI setup consisting of a piston connected to a pull out stub. To perform a test, air pressure is transmitted to the piston which is placed over the pull stub and screwed on the reaction plate. The air pressure induces an airtight seal formed between the piston gasket and the aggregate surface. When the pressure in the piston exceeds the cohesive strength of the binder or the adhesive strength of the binder-aggregate interface, a failure occurs. The pressure at failure is recorded and converted into the pull-off tensile strength (kPa).

70 AH GREYLING

December 2012


Figure 7-2- Cross section schematic of PATTI piston attached to a pull stub (Youtcheff & Aurilio, 1999)

The BiTVal report notes the following concerns related to the original Patti Test: “ 

Results could be significantly improved by including a temperature controlling system and a standard methodology for detecting the type of failure of the specimens.



Attention needs to be paid to examining the entire soak-time/pull-off strength curve while evaluating binder.



Little attention is given to the preparation and characterization of substrates other that glass.



Questions concerning the surface chemistry or roughness of substrates remain unanswered. “

In South Africa two types of pull-out test is specified in the (Asphalt Academy-TG1, 2007) 

Method MB-8: Pull out test method for surfacing aggregate



Method MB-9: Pliers test for assessment of adhesion properties

Both these tests are extremely complicated and difficult to complete and the results are almost entirely operator dependant. The test has low repeatability and is therefore not often used in practise.

71 AH GREYLING

December 2012


7.3 PROPOSED TEST METHOD 7.3.1 PATTI TEST The bitumen industry first utilized the Pneumatic Adhesive Tensile Testing Instrument, or PATTI, in the late 1990’s to evaluate adhesive loss of binder-aggregate systems exposed to moisture conditioning. (Youtcheff & Aurilio, 1999) & (Kanitpong & Bahia, 2003) . Initial tests identified several significant effects, including variations in preparing the test assembly (operator dependence), binder film thickness, and curing and testing temperatures.

Recent generations of the PATTI, notably the PATTI Quantum Gold (PQG), address some of these shortcomings while ensuring compliance with surface seal industry requirements (Miller, 2010). Further modifications and the development of a new Bitumen Bond Strength (BBS) Testing procedure (AASHTO TP91 , 2011) made the testing of the bitumen aggregate bond a reality.

Early generations of the PATTI consisted of a pressure hose, adhesion tester, piston, reaction plate and a metal pull-out stub. Figure 7-3 below shows a typical assembly.

PATTI

Reaction Plate

Pull-out Stub Substrate Piston

Figure 7-3- Original PATTI assembly.

72 AH GREYLING

December 2012


The original test procedure consisted of the following steps: 

A hot bitumen sample is applied to a glass substrate and allowed to cure for a fixed time interval.



A metal pull-out stub is applied to the bitumen sample and allowed to set for a given time interval.



After placing the piston over the pull-out stub, the reaction plate is fixed to the stub.



The pressure hose introduces compressed air to the piston, resulting in an upward force on the specimen and eventual failure of the binder. Failure occurs when the applied pressure exceeds the cohesive strength of the binder or the adhesive strength of the binder-aggregate interface.



The pressure at failure is recorded and the procedure is repeated for other test specimens. (Youtcheff & Aurilio, 1999)

While the original test procedure did yield quantitative information related to bond strength characteristics and failure behavior, research at University of Wisconsin-Madison (UWM) and University of Ancona – Italy (UAI) identified several factors influencing the effectiveness of the test method. Some of these factors include:



The binder film thickness between the stub and substrate could not be controlled easily.



Because the original pull-out stub measured only 12.7 mm in diameter, the stub geometry limited the measurement of smaller tensile strengths. Recent modifications to the pull-out stub design at UWM and UAI improved the geometry by nearly doubling the stub diameter.



The device did not record pressure over time, making the calculation of loading rate difficult. Rather, the PATTI reported only the real-time applied load but not within a computer-based graphical user interface.



The loading rate varied and could not be set easily. While the PATTI is equipped with a rate control dial, the dial did not effectively control the loading rate or report the real-time loading rate.



Initial tests were performed on glass substrates, hardly a suitable surface seal material.

7.3.2 INITIAL TEST METHOD LIMITATIONS & MODIFICATIONS The original PATTI test had limitations which were addressed in various ways. Improvements to the pull-out stub design, loading rate control and substrate preparation procedures represented significant advancements from the original PATTI test to its current form as the BBS test method. Binder film thickness is effectively controlled in the BBS test with an improved stub design. Loading rate is effectively controlled with new functions of the PATTI Quantum Gold.

Substrate surface characteristics are controlled with improved

substrate preparation procedures. Each of these improvements will be discussed in more detail in the sections below. 73 AH GREYLING

December 2012


DEVELOPMENT, TESTING AND EVALUATION OF RESULTS 8

DEVELOPMENT OF THE BITUMEN BOND STRENGTH TEST

8.1 INTRODUCTION Throughout the literature review the study focused on chip seals, their behaviour and especially the different factors that influence the bitumen binder and aggregate adhesion. The literature review also aimed to motivate the need for an adhesion test that will be capable of measuring the physical tensile strength required to break the bitumen and aggregate bond. The section below will aim to summarise the role that the University of Stellenbosch played in the development of the Bitumen Bond Strength Test procedure, using the PATTI Quantum Gold testing equipment, as well as to practically confirm some of the influences identified during the literature review.

8.2 UNIVERSITY OF STELLENBOSCH INVOLVEMENT The University of Stellenbosch (US) was requested by the University of Wisconsin Madison (UWM) in 2008 to assist the Asphalt Research Consortium (ARC) with the Federal Highway Administration (FHWA) Seal research project , and in particular to form part of the development of the proposed pull-off test. At this stage the need for an Adhesion Test was already identified and the UWM was in the process to search for suitable equipment. The US brief was to assist the UWM in the following: 

Assist in identifying and evaluating suitable equipment.



Assist in developing a testing protocol.



Evaluate the test for practicality and repeatability once complete.



To assist with inputs gained through experience on chip seals.

During the period from 2008-2010 the US completed these tasks and played an important role in the research and process that led to the final submission and approval of the Standard Method of Test for Determining Asphalt Binder Bond Strength by Means of the Asphalt Bond Strength (ABS) AASHTO TP 91-2011 as included in Addendum 1 for information purposes. This was achieved by the following actions: 

A study tour was undertaken to UWM by Professor Kim Jenkins and the author during August and September 2008. A report was completed by the author as part of this study and it is included as Addendum 2 for information purposes. 74

AH GREYLING

December 2012




Following the successful study tour a Research Group was established in January 2009 consisting of representatives from the US and UWM.



The research group completed parallel literature reviews and tests and combined their knowledge and experiences in monthly conference calls during 2009. The minutes of these conference calls as well as reports completed during this phase are in included as Addendum 3 for information purposes.



Mr Bryce Constable completed his Final Year Thesis for completion of his B.Eng. degree in 2009. He completed the majority of the first round testing and was mentored by the Author.



The Author, with the assistance of Mr Constable, completed a second round of testing to evaluate the Bitumen Bond Strength Test procedure. The results are included as Addendum 4 for reference.



Mr Timothy Miller completing his Master Thesis titled Development of Bond Strength Test for Improved Characterization of Asphalt Emulsions in 2010. This was based on the work done at the UWM and US.



The results of Millers’ study were accepted for presentation at the 2nd International Spray Sealing Conference in Australia. In 2010. It was included in the proceeding as follows: o

The Development of a Test Method for determining Emulsion Bond Strength using the Bitumen Bond Strength (BBS) Test by Timothy Miller , Andre Greyling , Prof Hussain Bahia & Prof Kim Jenkins

o 

The paper is included as Addendum 5 for information purposes.

Further work done by the Author lead to the acceptance of the study as a paper at the 10th Conference on Asphalt Pavements for Southern Africa in 2011 – It was published under the following title: o

The Development of a Test Method for determining Emulsion Bond Strength using the Bitumen Bond Strength (BBS) Test- A South African Perspective by Andre Greyling, Timothy Miller, Prof Kim Jenkins & Prof Hussain Bahia

o

The paper is included as Addendum 6 for information purposes.

The results of the testing completed during 2009 & 2010, as well as test performed by Mr Miller, forms the basis of the test results discussed in the sections below.

75 AH GREYLING

December 2012


8.3 MODIFICATION TO ORIGINAL PATTI TEST 8.3.1 INTRODUCTION The previous section touched briefly on the development of the original PATTI test into the formal Bitumen Bond Strength Test (BBS) procedure. To make this possible various shortcomings had to be addressed. The section below will aim to give a brief overview of these actions:

8.3.2 FILM THICKNESS Previous research completed at the UWM, identified film thickness as a critical parameter in investigating pull-off behavior. (Youtcheff & Aurilio, 1999) (Miller, 2010). Varying the binder film thickness will lead to variation in adhesion strength.

Early experimentation by Youtcheff and Aurilio to evaluate moisture

sensitivity utilized glass beads of 200 μm diameter mixed with bituminous binder to control film thickness. This had obvious shortcomings and was only used as a temporary solution.

With input from UAI, Kanitpong and Bahia further modified the pull-out stubs to better control binder film thickness (Canestrari, et al., 2010). They proposed using a smooth-surface aluminum stub and two metal support blocks to replace the glass beads. Photo 8-1 shows the modified stubs with metal support blocks, and Photo 8-2 depicts a later iteration of the pull-out stubs with aluminum frame supports.

Photo 8-1- Modified pull-out stubs with metal support blocks (Kanitpong & Bahia, 2003)

76 AH GREYLING

December 2012


In early rounds of experimentation, it became clear that film thickness was still not adequately controlled with the pull-out stub support system shown in Photo 8-2. Photo 8-3 shows an improved pull-out stub designed at UWM in conjunction with UAI.

Improvements to the original stub include an increase in stub diameter to 22 mm to increase the contact area and the addition of circumferential support edges to limit the vertical position of the stub surface and therefore the film thickness to 0.08mm. The Perimetrical channels in the stub edge allow excess binder to flow out from beneath the stub surface.

Photo 8-2- Modified pull-out stub with aluminum frame supports.

Photo 8-3- Modified BBS test pull-out stubs. 77 AH GREYLING

December 2012


8.3.3 LOADING RATE Bitumen’s viscoelastic nature necessitates effective loading rate control for the consistent evaluation of the pull-off tensile strength. Research by Meng confirmed that load control is critical for consistent pull-off tensile test results (Meng, 2010). Early versions of PATTI were found to inadequately control loading rate so at the beginning of 2009, SEMicro launched the PATTI Quantum Gold© (PQG) test instrument that incorporated user feedback into the revised design, including improved loading rate control. The ability to control the loading rate with a graduated rate control dial further improves consistency. The PQG comes equipped with LabView© software and effectively captures load over time, allowing for calculation of the loading rate.

8.3.4 SUBSTRATE & SURFACE ROUGHNESS As glass is only relevant as a control subgrade it was necessary to find a way to test the bitumen binder applied directly on aggregate as in practice. This was made possible by the improved substrate preparation procedures that involved the use of aggregate substrates that are actually used in practice. However, crushed aggregates commonly used in surface seals are not suitable substrate materials due to variations in in shape, surface roughness and texture. A procedure developed at UWM to prepare aggregate plates involves cutting large rocks into flat plates (Miller, 2010). The aggregate plates and disks are then lapped with a silicon carbide compound to achieve a consistent surface texture. While aggregate plates and disks may not fully capture aggregate surface characteristics, they represent substantial improvements over glass plates, which are still used as control surfaces. The smooth aggregate surfaces are seen and treated as a worst case scenario.

8.4 TEST APPARATUS AND SETUP 8.4.1 INTRODUCTION The culmination of the various research and test led to the establishment of the proposed Bitumen Bond Strength Test procedure (AASHTO TP-91 , 2011). The final procedure is attached in Addendum 1. The section below will aim to describe the proposed procedure on a step by step basis using the PQG with the modifications discussed in the sections above.

78 AH GREYLING

December 2012


8.4.2 STEP 1 – AGGREGATE PREPARATION The first requirement of the BBS Test is solid and flat aggregate subgrade samples of suitable size to be functional for testing. These samples are prepared by cutting pieces of large rock into flat plate for use as aggregate testing samples. The UWM originally proposed that a lapidary rock saw must be used to cut these aggregate plates. This type of saw is preferred over other, quicker methods, as it produces rock plates with smoother/parallel faces. (Constable, 2009) This is due to a screw-driven mechanism which slowly feeds the rock into the saw blade and can take about 45 minutes to complete one cut. The lapidary rock saw and screwdriven mechanism can be seen in Photo 8-4 & 8-5 below:

Photo 8-4- Rock Slab Saw (Miller, 2010)

Photo 8-5- Rock Slab Saw and Screw Driven Mechanism. (Miller, 2010) 79 AH GREYLING

December 2012


Due to the specialist nature of this equipment there were unfortunately no rock slab saw available at the University of Stellenbosch or in the greater Cape Town area. Initially rock samples where saw cut using a concrete saw but the aggregate samples were too large and uneven due to the manual feed of these saws. Please note the saw marks on the sawn slabs in Photo 8.9 below. The problem was solved at the US by sawing smaller geological ‘cores’ using a standard lab concrete saw. These cores are easier to saw and therefore give more uniform and smooth surface finishes.

8.4.3 STEP 2- AGGREGATE LAPPING Once the cores/plates have been cut it requires a form of smoothing/lapping to ensure a uniform surface roughness over various aggregates. Samples were lapped on both sides using a 280 Grit Silicon Carbide material. Future research should include for the measurement of the surface texture using a laser profilometer. For the sake of this study it was assumed that the smoothed lapped surface where all equal in texture. The lapping machine and grit may be seen in Photo 8-6 and Photo 8-7 respectively.

Photo 8-6- Lapping Machine (Constable, 2009)

80 AH GREYLING

December 2012


Photo 8-7- Ceramic Grit 280 (Constable, 2009)

8.4.4 STEP 3- CLEANING OF AGGREGATE Once plates have been lapped, they should be cleaned to remove any fine particles embedded within the plate surface. To clean the samples an ultrasonic cleaner was used. It is specified that the plates should be immersed in distilled/clean water within the ultrasonic cleaner for 60 minutes at a temperature of 60°C. Thereafter the plates can be hand dried. An ultrasonic cleaner is shown in Photo 8-8 below.

Photo 8-8- Ultrasonic cleaner 81 AH GREYLING

December 2012


8.4.5 STEP 4 – HEATING OF EMULSION Once the rock samples are complete and dried the required emulsion is placed in a plastic or another nonmetallic container and preheated in a forced-draft oven to an application temperature of 60°C. The emulsion should be covered to ensure that a dry top layer/skin does not form. The oven should always be pre heated and the sampled should not be heated for longer than one hour to avoid premature breaking.

8.4.6 STEP 5 HEATING OF AGGREGATE At the same time as heating the emulsion, the aggregate samples can be heated in a second forced-draft oven to a desired application temperature of 25°C. A laser thermometer should be used to ensure that the plate surface temperature is heated accurate enough. The required specification is 25°C +/- 2°C. Two rock plates showing saw marks and two cores are shown in Photo 8-9 below:

Photo 8-9- Granite and Tillite prepared aggregate samples

8.4.7 STEP 6-EMULSION ON AGGREGATE SAMPLE Silicone moulds are pre-prepared and should measure 50mm x 50mm (or at least 50mm in diameter to cover the PATTI base), with a 20mm diameter hole in the centre (for emulsion and stub application) and a thickness of at least 0.8 mm. After preheating the plates and emulsion, the silicone moulds are placed on the aggregate 82 AH GREYLING

December 2012


surface and liquid emulsion is placed into the silicone moulds with the use of a syringe or medicine dropper. The exact amount of emulsion is not critical as the film thickness will be controlled at 0.8mm by the modified stub.

Photo 8-10- Bitumen emulsion poured into silicone moulds

8.4.8 STEP 7 – CURING OF SPECIMEN The aggregate plate/silicone mold/emulsion combination (Photo 8-10) is placed into an environmental chamber or forced draft oven and allowed to cure under controlled conditions for specific curing intervals. During this time, the modified stainless steel pull-out stubs are preheated to a temperature of 60°C in a forceddraft oven. These aggregate plate/emulsion samples shall be tested at intervals of 2, 6 and 24 hours.

Photo 8-11- Environmental Chamber

83 AH GREYLING

December 2012


8.4.9 STEP 8 -APPLYING THE PULL-OUT STUB At each interval, 2, 6 and 24 hours, the necessary samples are removed from the forced-draft oven and the silicone molds encircling the emulsion are removed from the aggregate plate. The preheated pull-out stubs are then firmly placed on the cured emulsion ensuring that all four rims of the stub are touching the plate surface. Please note the gaps in the edges allowing excess binder to escape. The modified pull-out stubs, as described previously, can be seen in Photo 8-3 above:

8.4.10 STEP 9 -SAMPLE PREPARATION Once the pull-out stubs have been applied, the aggregate plate/emulsion/stub assembly is placed back into a forced-draft oven at a temperature of 25°C for approximately one and a half hours. This additional time allows the closed sample to acclimate to testing conditions. One can see the aggregate plate/emulsion/stub assembly in Photo 8.12 below;

Photo 8-12- Core Aggregate samples in Forced Draft Oven (Constable, 2009)

8.4.11 STEP 10 -TESTING After 1 hour the samples are removed from the oven and the excess emulsion on the plate surface is removed with a flat knife so that the emulsion along the edges of the stub does not influence the testing results. The PATTI Quantum Gold device is then used to test these samples. Small metal supports are used as a base for the pneumatic pressure ring around the stub. Special care should be taken to not disturb emulsion/stub bond the while affixing the pneumatic pull of ring. The photo below shows the PATTI Quantum Gold device set up and ready for a test: 84 AH GREYLING

December 2012


Photo 8-13- PATTI Quantum Gold with testing assembly (Constable, 2009) Plate

The PATTI unit is linked to a computer via a data cable where it is connected with LabView Software© The results are displayed visually and the data is stored to a text file. Compressed air or CO2 canisters are used to deliver the required pressure force. A gauge on the front of the equipment indicates the supply pressure. The loading rate can be adjusted using a dial mounted on top of the testing unit. A top view of the PATTI unit and loading rate dial can be seen in the two photos below:

Photo 8-14- Top View of PATTI Quantum Gold

85 AH GREYLING

December 2012


Photo 8-15- Loading Rate dial

The PATTI Quantum Gold software package allows the user to select various stub designs and other testing parameters. A real-time display of the test (time vs. tensile strength), compliance with ASTM standards and deliverance of spread sheet outputs are all featured options. (Constable, 2009) A screen view of the software interface can be seen in Figure 8-1. The test results as well as the screenshot report are saved and can be accessed at any time for further calculations and record purposes

. Figure 8-1- PATTI Quantum Gold user interface

86 AH GREYLING

December 2012


8.5 GENERAL COMMENTS As part of the evaluation of the testing procedure the following valuable lessons were learned and the following advice and comments should be noted: 

Plates have to be perfectly smooth and flat otherwise emulsion soaked underneath the silicone moulds.



Not all newly manufactured stubs had smooth threads and this required some manual force to screw in the pressure ring. This led to stubs losing adhesion before the tests could take place. Special care should be taken that the pressure ring and the pull out stub screw in perfectly.



The loading rate dial was difficult to control and very sensitive to adjustment. It was later discovered that the dial of the PQG used at the US was malfunctioning. After much trial and error it was possible to achieve application rates of between 500 – 1000 kPa/s.



Without a rock plate saw it is better to use geological cores and saw them into disks.

87 AH GREYLING

December 2012


9

EXPERIMENTAL SETUP

9.1 INTRODUCTION Over 200 tests where completed over a two year period at the University of Stellenbosch. Mr Bryce Constable completed a first set of test in 2009 for the completion of his thesis. These tests were done at a curing temperature of 25°C and not at 30°C as the formal procedure was not finalised yet. A second more detailed test set was completed in 2010 at 30°C by Mr Constable under supervision of the Author. This made it possible to correlate the results for the US with the UWM.

The section below will discuss these experimental setups and results in more detail.

9.2 MATERIALS 9.2.1 EMULSIONS The following four emulsion types where evaluated during the testing phase:

1. Cationic Rapid Set Emulsion with 65% Bitumen content (CRS65) 2. Anionic Slow Set Emulsion with 60% Bitumen content (SS60) 3. Cationic Rapid Set Emulsion with 65% Bitumen and Modified with 3% Latex (CRS65+3%) 4. Anionic Rapid Slow Set Emulsion with 60% Bitumen and Modified with 3% Latex (SS60+3%)

This combination of emulsion made it possible to evaluate Anionic vs. Cationic as well as Modified vs. Unmodified emulsion adhesion development. The Emulsion used where supplied courtesy of COLAS and TOSAS South Africa.

9.2.2 AGGREGATES Two sets of aggregate samples from quarries previously used for chip seal construction where used for testing. Granite and Tillite aggregate cores where supplied by the South African National Road Agency –Western Region. Granite is an Acidic Igneous Rock and Tillite is a slightly less Acidic but slightly more porous Sedimentary rock type.

88 AH GREYLING

December 2012


9.3 EXPECTED RESULTS Based on the theoretical information gathered in the literature review section the following test results can be expected:

1. Adhesion will increase with curing time for all combinations of variables. 2. Cationic Emulsion will develop stronger adhesion than Anionic emulsion due to the fact that the positive charged cationic emulsion forms stronger bonds with the negatively charged acidic rock types used. 3. The stable grade Anionic Emulsion will show slower adhesion development as the stable grade emulsion is expected to have a longer breaking process. 4. The rapid setting Cationic Emulsion will develop strength earlier than the Anionic Emulsion as it is expected from a rapid setting emulsion. 5. The modification of the Emulsion with Latex should lead to increased adhesion as it is expected from modified binders. 6. Granite is slightly more acidic and coarser than tillite and will therefore develop stronger adhesion with the cationic emulsions. 7. Tillite is slightly less acidic and more porous that granite and should therefore develop stronger adhesion with the anionic emulsion than with granite.

If the BBS test procedure is suitable for testing adhesion development it will confirm the expected results above.

9.4 EXPERIMENTAL SETUP As part of the test procedure evaluation a total of 198 tests where completed during eleven test sets shown in Table 9-1 below. Each test set was prepared to consist of the following combinations:



One emulsion type , either CRS65 , CRS65+3% Latex, SS60 or SS60+3% Latex



Two aggregate types, Granite of Tillite



Three curing times , 2,6 and 24 hours



At least three stubs for each combination. (In most cases four to five stubs was used to ensure that the test yielded three valid results per combination.)

89 AH GREYLING

December 2012


During the different test sets the curing temperature as well as the pull of rate was varied to evaluate the influence of the variability. The results of the various combinations is shown and discussed in the section below.

Table 9-1- Experimental Setup and Test Sets completed. Note

5 SETS of CRS65

2 SETS of CRS65+3 % Latex

2 SETS of SS60

2 SETS of SS60+3% Latex

Tests Set Number 1 2 3 4 5 1 2 3 4 5 6 7 6 7 8 9 8 9 10 11 10 11

Emulsion

Aggregate

CRS65 CRS65 CRS65 CRS65 CRS65 CRS65 CRS65 CRS65 CRS65 CRS65 CRS65+3%Latex CRS65+3%Latex CRS65+3%Latex CRS65+3%Latex SS60 SS60 SS60 SS60 SS60+3%Latex SS60+3%Latex SS60+3%Latex SS60+3%Latex

Granite Granite Granite Granite Granite Tillite Tillite Tillite Tillite Tillite Granite Granite Tillite Tillite Granite Granite Tillite Tillite Granite Granite Tillite Tillite

Curing Time (Hours) 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours 2,6& 24 hours

Curing Temp 25°C 25°C 30°C 30°C 30°C 25°C 25°C 30°C 30°C 30°C 25°C 30°C 25°C 30°C 30°C 30°C 30°C 30°C 30°C 30°C 30°C 30°C

Pull Off Rate (kPa) 300 300 700 800 900 300 300 700 800 900 400 950 400 950 900 700 900 700 800 700 800 700

9.5 TEST RESULTS- EMULSIONS 9.5.1 CATIONIC 65% EMULSION (CRS65) The results for the first five sets (Set 1-5) of the Cationic Rapid Set Emulsions that were tested are shown in Figure 9-1 below. Please note that the average values of at least three results per set were used for graphical representation purposes:

90 AH GREYLING

December 2012


CRS65 on GRANITE & TILLITE 1200

PULL-OUT TENSION ( kPa)

1000

TEST 1 @300kPa/s @ 25°C TEST 2 @300kPa/s@ 25°C TEST 3 @ 700kPa/s @ 30°C TEST 4@800kPa/s @ 30°C TEST 5 @900kPa @ 30°C

800

600

400

200

0

2 HOURS on 6 HOURS on 24 HOURS on 2 HOURS on 6 HOURS on 24 HOURS on GRANITE GRANITE GRANITE TILLITE TILLITE TILLITE CURING TIME

Figure 9-1- CRS65 Test Results on Granite & Tillite The following information can be deduced from Figure 9-1 above: 

Adhesion increased with time based on the higher results achieved after 2, 6 and 24 hours for both granite and tillite.



The results after 2 and 6 hours shows lower results for the two sets cured at 25°C and tested at 300kPa/s on both granite and tillite



The results after 24 hours shows higher results at the 25°C curing temperature and 300kPa/s pull of rate for the granite but results on tillite show limited variation.

The influences of the various factors will be discussed in consequent sections. Table 9-2 below shows a statistical analysis of the results of test Set 5. The first section shows the pull of rate in kPa/s and the second section shows the results. Note the Covariance values.

Positive Covariance values indicates that higher than average values of one variable tend to be paired with higher than average values of the other variable. Negative Covariance indicates that higher than average values of one variable tend to be paired with lower than average values of the other variable. Zero Covariance indicates that two random variables are independent. However, a covariance of zero does not necessarily mean that the variables are independent. A nonlinear relationship can exist that still will result in a covariance value of zero. Covariance values of less than 10% usually indicate good repeatability. 91 AH GREYLING

December 2012


Table 9-2- Covariance Analysis of CRS Test Set 5 on Granite Granite CRS65

2H

CRS65

6H

CRS65

24H

Granite CRS65

2H

CRS65

6H

CRS65

24H

CRS65 on GRANITE TEST SET 5 Rep Rate (kPa/s) St Dev. St Error 1 869.92 2 860.52 867.67 6.33 3.66 3 872.57 1 1092.62 2 1134.89 1125.19 28.97 16.73 3 1148.07 1 909.09 2 985.87 930.80 48.05 27.74 3 897.44 Rep Tension (kPa) StDev St Error 1 427.33 2 427.33 424.62 4.69 2.71 3 419.20 1 511.34 2 549.29 533.02 19.54 11.28 3 538.45 1 866.36 2 939.53 919.66 46.65 26.93 3 953.08

COV 0.73%

2.57%

5.16% COV 1.11%

3.67%

5.07%

Table 9-3- Covariance Analysis of CRS Test Set 5 on Tillite Tillite CRS65

2H

CRS65

6H

CRS65

24H

Tillite CRS65

2H

CRS65

6H

CRS65

24H

CRS65 on TILLITE TEST SET 5 Rep Rate (kPa/s) St Dev St Error 1 855.19 2 810.13 830.52 22.84 13.18 3 826.24 1 925.23 2 977.95 973.35 46.00 26.56 3 1016.87 1 860.75 2 951.41 918.46 50.15 28.95 3 943.21 Rep Tension (kPa) StDev St Error 1 400.23 2 394.81 406.55 15.88 9.17 3 424.62 1 766.09 2 687.50 765.19 77.24 44.60 3 841.97 1 820.29 2 787.77 835.65 57.13 32.98 3 898.88

COV 2.75%

4.73%

5.46% COV 3.91%

10.09%

6.84%

92 AH GREYLING

December 2012


9.5.2 ANIONIC 60% EMULSION (SS60) The average results for the three samples of each of the two test sets (Set 8&9) completed using Stable Grade Emulsion is shown in Figure 9-2 below:

SS60 on GRANITE & TILLITE 1200

SS60 @ 900 kPa/s @25C SS60 @ 700kPa/s @ 30C


1000

800

600

400

200

0 2 HOURS on GRANITE

6 HOURS on GRANITE

24 HOURS on GRANITE

2 HOURS on TILLITE

6 HOURS on TILLITE

24 HOURS on TILLITE

CURING TIME

Figure 9-2- SS60 Test Results on Granite & Tillite

The following information can be deduced from the figure above:



No adhesion increase over time can be seen for the SS60/Granite combination.



Limited adhesion increase over time can be seen for the SS60/Tillite combination.



Higher pull off rate at lower temperatures deliver slightly higher results and this indicate that the test results are more susceptible to pull off rate than curing temperature.

Please note the detailed results of test Set 8 completed in 2009 below. It is interesting to note that the tables show that the majority of the COV values are smaller than 10% in the 2 and 6 hour ranges. For the longer (24 hour) curing temperature the COV shows a much higher variability in the bond strength (tension) results.

93 AH GREYLING

December 2012


Table 9-3- Covariance Analysis of SS60 Test Set 8 on Granite Granite SS60

2H

SS60

6H

SS60

24H

Granite SS60

2H

SS60

6H

SS60

24H

SS60 on GRANITE- TEST SET 8 (2009) Rep Rate (kPa/s) St Dev St Error 1 839.45 2 823.72 820.50 20.75 11.98 3 798.32 1 784.61 2 789.18 782.65 7.70 4.44 3 774.16 1 696.53 2 658.42 690.24 29.18 16.85 3 715.76 Rep Tension (kPa) StDev St Error 1 497.79 2 476.11 482.44 13.37 7.72 3 473.40 1 470.69 2 467.98 466.18 5.64 3.26 3 459.85 1 495.08 2 473.40 545.67 106.95 61.75 3 668.53

COV 2.53%

0.98%

4.23% COV 2.77%

1.21%

19.60%

Table 9-4- Covariance Analysis of SS60 Test Set 8 on Tillite SS60 on TILLITE- TEST SET 8 (2009) Rep Rate (kPa/s) St Dev St Error 1 947.85 SS60 2H 2 905.31 941.39 33.32 19.24 3 971.01 1 770.90 SS60 6H 2 760.48 762.43 7.68 4.43 3 755.91 1 702.66 SS60 24H 2 692.34 747.14 86.13 49.73 3 846.41 Tillite Rep Tension (kPa) StDev St Error 1 443.59 SS60 2H 2 438.17 445.40 8.28 4.78 3 454.43 1 457.14 SS60 6H 2 451.72 459.85 9.77 5.64 3 470.69 1 524.89 SS60 24H 2 625.17 568.26 51.49 29.73 3 554.71 Tillite

COV 3.54%

1.01%

11.53% COV 1.86%

2.12%

9.06%

94 AH GREYLING

December 2012


9.5.3 CATIONIC 65% +3% LATEX MODIFIED EMULSION The average results for the three samples of each of the two test sets (Set 6&7) completed using Cationic Emulsion Modified with 3% Latex shown in Figure 9-3 below:

CRS65+3%Latex on GRANITE & TILLITE 1200

CRS65+3% @ 400 kPa/s (25C) CRS65+3% @ 950kPa/s (30C)


1000

800

600

400

200


6 HOURS on GRANITE

24 HOURS on GRANITE

2 HOURS on TILLITE

6 HOURS on TILLITE

24 HOURS on TILLITE

CURING TIME

Figure 9-3- CRS65+3% Latex - Test Results on Granite & Tillite

The following information can be deduced from the figure above: 

A clear adhesion increase over time can be seen for the CRS65+3% Latex /Granite combination.



A clear adhesion increase over time can be seen for the CRS65+3% Latex /Tillite combination.



Higher pull off rate at higher temperatures deliver slightly higher results.



It again seems that the test results are more susceptible to the pull off rate than curing temperature.



There is little difference in adhesion development between Granite and Tillite with Tillite showing marginally higher results after 24 hours.

Please note the detailed results of test Set 7 completed in 2010 below. Please again note the higher Variance in both rate of applied pull off tension as well as the results after 24 hours.

95 AH GREYLING

December 2012


Table 9-5- Covariance Analysis of CRS65+3% Latex Test Set 7 on Granite CRS65+3% LATEX on GRANITE – TEST SET 7 Granite Rep Rate (kPa/s) St Dev St Error 1 617.82 CRS65+3% LATEX 2H 2 738.91 696.61 68.30 39.43 3 733.10 1 1010.33 CRS65+3% LATEX 6H 2 977.60 952.40 73.82 42.62 3 869.28 1 974.32 CRS65+3% LATEX 24H 2 944.46 960.51 15.06 8.69 3 962.75 Granite Rep Tension (kPa) StDev St Error 1 443.59 CRS65+3% LATEX 2H 2 438.17 439.08 4.14 2.39 3 435.46 1 836.55 CRS65+3% LATEX 6H 2 809.45 824.81 13.91 8.03 3 828.42 1 806.74 CRS65+3% LATEX 24H 2 1018.13 989.22 169.88 98.08 3 1142.79

COV 9.80%

7.75%

1.57% COV 0.94%

1.69%

17.17%

Table 9-6- Covariance Analysis of CRS65+3% Latex Test Set 7 on Tillite CRS65+3% LATEX on TILLITE-– TEST SET 7 Rep Rate (kPa/s) St Dev St Error 1 957.38 CRS65+3% LATEX 2H 2 899.60 944.59 40.16 23.19 3 976.80 1 967.78 CRS65+3% LATEX 6H 2 993.96 974.01 17.69 10.21 3 960.28 1 742.96 CRS65+3% LATEX 24H 2 881.36 835.22 79.90 46.13 3 881.36 Tillite Rep Tension (kPa) StDev St Error 1 449.01 CRS65+3% LATEX 2H 2 432.75 446.30 12.42 7.17 3 457.14 1 801.32 CRS65+3% LATEX 6H 2 823.00 846.49 60.44 34.89 3 915.14 1 974.77 CRS65+3% LATEX 24H 2 1156.34 1095.81 104.83 60.52 3 1156.34 Tillite

COV 4.25%

1.82%

9.57% COV 2.78%

7.14%

9.57%

96 AH GREYLING

December 2012


9.5.4 ANIONIC% +3% LATEX MODIFIED EMULSION The average results for the three samples of each of the two test sets (Set 10&11) completed using Anionic Slow Set Emulsion Modified with 3% Latex shown in Figure 9-4 below:

SS60+3% LATEX on GRANITE & TILLITE 1200

SS60+3% @ 700kPa/s (30C) SS60+3% @ 800 kPa/s (30C)


1000

800

600

400

200


6 HOURS on GRANITE

24 HOURS on GRANITE

2 HOURS on TILLITE

6 HOURS on TILLITE

24 HOURS on TILLITE

CURING TIME

Figure 9-4- SS60+3% Latex - Test Results on Granite & Tillite

The following information can be deduced from the figure above: 

There is no clear increase in adhesion development of the first 24 hours with the SS60+3% Latex /Granite combination. This is expected from a Slow Setting Anionic Emulsion on an Acidic aggregate.



The SS60+3% Latex /Tillite combination show limited initial adhesion but increased adhesion after 24 hours. This can be attributed to the increased porosity of the sedimentary tillite aggregate.



There is limited sensitivity between a pull off rate of 700kPa/s and 800kPa/s.



The figure also shows the need for the investigation of longer curing times.

Please note the detailed results of test Set 10 completed in 2010 on the next page. 97 AH GREYLING

December 2012


Table 9-7- Covariance Analysis of SS60+3% Latex Test Set 10 on Granite SS60+3% LATEX on GRANITE-TEST SET 10 Granite Rep Rate (kPa/s) St Dev St Error 1 665.96 SS60+3% LATEX 2H 2 673.40 665.78 7.71 4.45 3 657.98 1 627.38 SS60+3% LATEX 6H 2 607.75 609.93 16.47 9.51 3 594.65 1 1015.17 SS60+3% LATEX 24H 2 1000.50 993.63 25.67 14.82 3 965.22 Granite Rep Tension (kPa) StDev St Error 1 478.82 SS60+3% LATEX 2H 2 473.40 472.50 6.82 3.94 3 465.27 1 519.47 SS60+3% LATEX 6H 2 503.21 505.92 12.42 7.17 3 495.08 1 476.11 SS60+3% LATEX 24H 2 484.24 477.92 5.64 3.26 3 473.40

COV 1.16%

2.70%

2.58% COV 1.44%

2.45%

1.18%

Table 9-8- Covariance Analysis of SS60+3% Latex Test Set 10 on Tillite SS60+3% LATEX on TILLITE-TEST SET 10 Rep Rate (kPa/s) St Dev St Error 1 755.91 SS60+3% LATEX 2H 2 748.05 750.67 4.54 2.62 3 748.05 1 816.60 SS60+3% LATEX 6H 2 806.10 807.01 9.17 5.30 3 798.32 1 941.11 SS60+3% LATEX 24H 2 966.38 965.36 23.76 13.72 3 988.60 Tillite Rep Tension (kPa) StDev St Error 1 457.14 SS60+3% LATEX 2H 2 449.01 449.92 6.82 3.94 3 443.59 1 486.95 SS60+3% LATEX 6H 2 484.24 484.24 2.71 1.56 3 481.53 1 676.66 SS60+3% LATEX 24H 2 776.93 717.31 52.76 30.46 3 698.34 Tillite

COV 0.61%

1.14%

2.46% COV 1.52%

0.56%

7.36%

98 AH GREYLING

December 2012


9.6 CONFIRMING SIGNIFICANT FACTORS INFLUENCING BOND STRENGTH DEVELOPMENT Miller did some sterling work investigated factors critical to bond strength development (Miller, 2010). The factors investigated included various substrate types, moisture condition, surface roughness, loading rate, curing temperature and curing humidity. An analysis of variance (ANOVA) for his screening experiments is shown in Table 9-9 below.

His experiments, completed with a combination of data form UWM and the US identified that the loading rate, the curing temperature and humidity, and the substrate type has significant influence on the bond strength development. The results indicate that the loading rate is the most significantly influence on the pull-out tension response. Other factors investigated included substrate and binder type as well as curing variables related to temperature, humidity and time (Miller, 2010).

Table 9-9- Analysis of variance for the factor screening experiment. (Miller, 2010) Source

DF

Seq. SS

Adj. SS

Adj. MS

F

P

Substrate

1

1051.6

1051.6

1051.6

8.2

0.006

Moisture

1

0.5

0.5

0.5

0.00

0.952

Roughness

1

90.9

90.9

90.9

0.7

0.404

Loading Rate

1

31926.2

31926.2

31926.2

248.5

0.000

Temperature

1

1967.0

1967.0

1967.0

15.3

0.000

Humidity

1

1105.3

1105.3

1105.3

8.6

0.005

Error

57

7322.4

7322.4

128.5

Total

63

43463.8

S = 11.3342

R-Sq = 83.15%

R-Sq (adj) = 81.38%

9.6.1 LOADING RATE EXPERIMENT Loading rate was further identified as a factor that significantly influencing the pull-out tension response. This is primarily due to the visco-elastic properties of bitumen discussed previously. Miller did extensive testing and proved that a power law model adequately captures the relationship between loading rate and pull-out tension. The results in Figure 9-5 show that loading rates between 690-1030 kPa/s appear to exhibit a linear relationship above pull-out tension values of 690 kPa.

99 AH GREYLING

December 2012


Loading rates exceeding 2700 kPa/s lead to increasing variability in both the pull-out tension values and loading rate. Therefore, loading rates exceeding 2700 kPa/s should be avoided to minimize experimental error in order to obtain valid results.

Figure 9-5- Loading rate and pull-out tension are described by a power law model. (Miller, 2010)

Some of the test done by the US confirmed these results. Figure 9-6 shows the influence of loading rates on two different aggregates and one emulsion combinations. The figure clearly indicates that there is an increase in the pull out tension as the rate of application increases. The results are however limited to the 400 kPa/s to 950 kPa/s range.

100 AH GREYLING

December 2012


CRS65+3%Latex on GRANITE & TILLITE 1200 CRS65+3% @ 400 kPa/s CRS65+3% @ 950kPa/s


1000

800

600

400

200


6 HOURS on GRANITE

24 HOURS on GRANITE

2 HOURS on TILLITE

6 HOURS on TILLITE

24 HOURS on TILLITE

CURING TIME

Figure 9-6- Loading rate and pull-out tension results US

9.6.2 CURING CONDITIONS EXPERIMENT It is expected that the effect of curing temperature and humidity will play a very important role in the bond strength development between bitumen and aggregate. Miller investigated the effects of curing temperature and humidity on pull-out tension (Miller, 2010) and confirmed the influences.

In his experiment, the loading rate was fixed at approximately 700 kPa/s for four curing conditions. Samples were cured in an environmental chamber at the prescribed curing intervals of 2, 6 and 24 hours. The experiment used a cationic rapid-setting emulsion with high viscosity (CRS-2), granite and limestone substrates, and the following curing conditions: 

Samples cured at 35 °C and 30 percent relative humidity.










101 AH GREYLING

December 2012


An ANOVA for the curing conditions experiment, shown in Table 9-10, indicates that aggregate type, curing temperature, and curing interval are statistically significant main effects at a 95% confidence level, while temperature-curing interval and humidity-curing interval interactive effects are also statistically significant at this confidence level. Other important results suggest that:



Significant strength gains were observed between two and six hours at the 35 °C temperature level.



Granite outperformed limestone in three of four curing conditions after six hours.



Humidity did not significantly affect the pull-out tension response.



Samples tested at 35 °C and 30 percent relative humidity performed better than samples tested at other curing conditions.



Samples exhibited only slight differences in the pull-out tension response after 24 hours of curing.

Table 9-10-Analysis of variance for the curing conditions experiment. (Miller, 2010) Source

DF

Seq. SS

Adj. SS

Adj. MS

F

P

Substrate

2

2862.9

2862.9

1431.4

16.44

0.000

Curing Temperature (°C)

1

2079.2

2079.2

2079.2

23.88

0.000

Curing Humidity (% RH)

1

60.6

60.6

60.6

0.70

0.417

Curing Interval

2

113689.5

113689.5

56844.8

652.96

0.000

Substrate-Temperature

2

247.0

247.0

123.5

1.42

0.271

Substrate-Humidity

2

559.2

559.2

279.6

3.21

0.067

Substrate-Interval

4

838.9

838.9

209.7

2.41

0.092

Temperature-Humidity

1

54.4

54.4

54.4

0.63

0.441

Temperature-Interval

2

3695.7

3695.7

1847.8

21.23

0.000

Humidity-Interval

2

1354.8

1354.8

677.4

7.78

0.004

Error

16

1392.9

1392.9

87.1

Total

35

126835.1

S = 9.33046

R-Sq = 98.9%

R-Sq (adj) = 97.6%

9.6.3 EMULSION TYPE EXPERIMENT Miller also developed a detailed materials experiment to investigate a variety of emulsion and substrate combinations (Miller, 2010) The experiment included five emulsion types and three substrate types at four curing intervals. Figure 9-7 and Figure 9-8 show results from this experiment.

102 AH GREYLING

December 2012


Two initial observations can be made in Figure 9-7: that both cationic rapid-setting emulsions (CRS-2 lab and CRS-2 field) perform better than polymer modified cationic rapid-setting emulsions (CRS-2P) and high-float anionic rapid setting (HFRS-2) emulsions; and that all emulsion types exhibit sharp increases in pull-out tension initially, with relative gains in tensile strength diminishing over time. A power law model adequately characterizes the relationship between curing interval and pull-out tension. In Figure 9-8, glass plates yield a near-perfect correlation using a power law model, with solid aggregate plates and chip substrates also yielding very strong relationships.

Figure 9-7- Pull-out tension values differ for various emulsion types at a range of curing intervals (Miller, 2010)

Figure 9-8- Pull-out tension values differ for various substrate types at a range of curing intervals (Miller, 2010) 103 AH GREYLING

December 2012


The US conducted test on granite and tillite aggregates using various modified and unmodified emulsion combinations. The results of these tests correlate well with the results from UWM and are shown in the Figure 9-9 below:

SS60 & CRS65 on GRANITE & TILLITE 1200

SS60 @ 900 kPa/s @25C SS60 @ 700kPa/s @ 30C CRS65 @ 900kPa/s @25C CRS65 @ 700kPa/s @ 30C


1000

800

600

400

200


6 HOURS on GRANITE

24 HOURS on GRANITE

2 HOURS on TILLITE

6 HOURS on TILLITE

24 HOURS on TILLITE

CURING TIME

Figure 9-9- Pull Out Tension Values for SS60& CRS65 on Granite & Tillite

Figure 9-9 shows a combination of test results for CRS65 and SS60 bitumen emulsions on granite and tillite aggregates for different loading rates. The following points of interest are noted: 

The CRS65-granite combination outperforms the CRS65-tillite combination at all curing intervals. The pull-out tension appears to increase linearly with curing time.



SS60 results show very little strength increase over 24 hours, except on tillite at 24 hours. The result is expected from a slow setting emulsion applied on acidic aggregates such as granite and tillite. These aggregates contain silica and have a strong negative charge in the presence of water. This negative charge attracts positively charged cationic bitumen particles, leading to destabilization of the surfactant system and subsequent coagulation of the bitumen particles. This breaking mechanism is absent when anionic emulsions are used with acidic aggregates (Louw, et al., 2004)



Tests conducted at 900 kPa/s show slight increases in pull–out tensions values over those tested at 700 kPa/s, thereby confirming that higher loading rates result in higher pull-out tension values. 104

AH GREYLING

December 2012


SS60+3%Latex & CRS65+3%Latex on GRANITE & TILLITE 1200

SS60+3% @ 800 kPa/s (30C) SS60+3% @ 700kPa/s (30C) CRS65+3% @ 400 kPa/s CRS65+3% @ 950kPa/s


1000

800

600

400

200


6 HOURS on GRANITE

24 HOURS on GRANITE

2 HOURS on TILLITE

6 HOURS on TILLITE

24 HOURS on TILLITE

CURING TIME

Figure 9-10- Pull Out Tension Values for SS60+3% Latex CRS65+3% Latex on Granite & Tillite

Figure 9-10 shows a combination of test results for CRS65 + 3% latex and SS60 + 3% latex on granite and tillite aggregate substrates. The following points of interest are noted: 

Tests performed at 950 kPa/s show higher results and confirm the findings of Miller that higher loading rates contribute to higher pull-out tension values.



After 2 hours curing time, results exhibit minimal differences in the results on all four emulsion/aggregate combinations.



After 6 hours the results for CRS65 + 3% latex/granite show higher tensile strength values than SS60 + 3% latex on granite.



After 6 hours, there are limited differences in the results of the CRS65 + 3% latex/tillite and SS60 + 3% latex/ tillite.



After 24 hours, the CRS65/granite combination shows significantly higher strength values than the SS60-granite combination but smaller values that the CRS65 + 3% latex/tillite and SS60+ 3 % latex/tillite combinations.



The modified emulsions performs better that the unmodified emulsions after 24 hours.

105 AH GREYLING

December 2012




The CRS65 + 3% latex/tillite and SS60 + 3% latex/tillite combinations values relate well and seem to be superior to granite.

CRS65 & CRS65+3%Latex on GRANITE & TILLITE 1200

CRS65 @ 700kPa/s @ 30C CRS65 @ 900kPa/s @25C CRS65+3% @ 400 kPa/s CRS65+3% @ 950kPa/s


1000

800

600

400

200


6 HOURS on GRANITE

24 HOURS on GRANITE

2 HOURS on TILLITE

6 HOURS on TILLITE

24 HOURS on TILLITE

CURING TIME

Figure 9-11- Pull Out Tension Values for CRS65& CRS65+3% Latex on Granite & Tillite

Figure 9-11 shows a combination of test results for CRS65 & CRS65 + 3% on granite and tillite aggregate substrates. The following points of interest are noted: 




The unmodified cationic emulsion (CRS65) outperforms the modified emulsion in after two and six hours on the tillite aggregate






After 6 hours the results for CRS65 + 3% latex/granite show higher tensile strength values than unmodified CRS65 on granite.



After 6 hours, the CRS65 on tillite shows higher results than the CRS65 + 3% on tillite.



After 24 hours, the CRS65-granite combination shows lower strength values than the CRS65+3% Latex/granite combination but smaller values that the CRS65-tillite and CRS65+ 3 % latex/tillite combinations. 106

AH GREYLING

December 2012




The modified emulsions perform better that the unmodified emulsions after 24 hours.



The CRS65 + 3% latex/tillite and CRS65 /tillite combinations values relate well and seem to be superior to granite.

SS60 & SS60+3% Latex on GRANITE & TILLITE 1200


1000

SS60 @ 700kPa/s @ 30C SS60 @ 900 kPa/s @25C SS60+3% @ 700kPa/s (30C) SS60+3% @ 800 kPa/s (30C)

800

600

400

200


6 HOURS on GRANITE

24 HOURS on GRANITE

2 HOURS on TILLITE

6 HOURS on TILLITE

24 HOURS on TILLITE

CURING TIME

Figure 9-12- Pull Out Tension Values for SS60& SS60+3% Latex on Granite & Tillite

Figure 9-12 shows a combination of test results for SS60 & SS60 + 3% on granite and tillite aggregate substrates. The following points of interest are noted: 

There is only a slight increase after 2, 6 and 24 hours on the SS60 Granite combination. This is expected from an Anionic Emulsion on an Acidic aggregate.



After 2 hour the Granite and SS60 & SS60+3% Latex combination shows slightly higher values that the SS60 & SS60+3% Latex and Tillite combinations.



After 6 hour the modified and unmodified emulsion shows comparable results on both granite and tillite.



After 24 hours the SS60 & SS60+3% Latex shows much higher result on tillite than on granite.

107 AH GREYLING

December 2012


9.7 CORRELATIONS OF BBS TEST RESULTS In order to validate the results of the BBS test procedure, Miller correlated the BBS test results to two other commonly used test procedures. To understand and quantify aggregate retention, Miller correlated BBS test results to results obtained using the ASTM D7000 Sweep test procedure (Miller, 2010; ASTM D7000, 2008).

In the sweep test, bitumen is applied at a fixed application rate to a felt disk before aggregate chips are applied and compacted. The test apparatus brushes the sample for 1 minute in an attempt to simulate the mechanical brooming action experienced by newly-constructed chip seals. The response variable considered in the sweep test is percent aggregate loss. 9.7.1.1 SWEEP TEST Establishing a correlation between BBS test results and sweep test results entailed comparing the pull-out tensile strength values obtained using the BBS test to aggregate loss measured using the sweep test at identical curing conditions. Given the existing recommended sweep test performance limit of 10 percent aggregate loss, Miller devised a test correlation between BBS and sweep test results. Sweep test samples prepared with granite and limestone aggregates and CRS-2 and CRS-2P emulsions are compared to BBS test results for similar material combinations, as shown in Figure 9-13.

At 2 hours curing, neither the sweep test samples nor the BBS test samples exhibit good performance in terms of aggregate retention or pull-out tension as indicated by high aggregate loss in the sweep test and low pullout tension values in the BBS test. At 6 hours curing, cohesion and adhesion become more evident with improved aggregate retention and pull-out tension values. After 24 hours curing, samples exhibit greater performance in both tests. As with other experiments, a power law model appears to adequately characterize the relationship between BBS results and sweep test results, with R2 > 0.995. Based on this relationship, a minimum BBS specification limit of 850 kPa may be suggested, a limit that signifies when the binder has gained sufficient bond strength to achieve less than 10 percent aggregate loss as measured by the sweep test.

During the various University Stellenbosch tests only the CRS65 and both the Modified Emulsion, cured for 24 hours, achieved the 850 kPa threshold on both granite and tillite. This is based on curing temperatures of 30°C. In practice road temperatures will in some areas exceed 60°C and a chip seal using emulsion as binder

108 AH GREYLING

December 2012


will usually only be opened after 24 hours. The values are therefore perceived reasonable and a good starting point for further research.

Figure 9-13- When sweep test results are compared to BBS test results, a potential BBS specification limit may be proposed at 850 kPa to define a specification target range (Miller,2010)

9.7.1.2 DYNAMIC SHEAR RHEOMETER (DSR) STRAIN SWEEP Strain sweep procedures developed for the dynamic shear rheometer (DSR) were also compared to BBS results (Miller, 2010). DSR strain sweep results were considered for two evaluation criteria. Test results were analyzed in the linear range (G* / sin  at 12 percent strain) and non-linear range (G* / sin  at 40 percent strain). Emulsion residues cured on granite and limestone substrates demonstrated the effect of curing temperature on strength gain, particularly at early curing intervals. None of the samples attained the full strength of the neat base binder, indicating the presence of water in the emulsion even after 24 hours of curing. BBS test results are correlated with emulsion residue properties in Figure 9-14. Results demonstrate a strong correlation between pull-out tensile strength and resistance to deformation (G* / sin δ) of the emulsion residue. The effect of curing time is also seen in comparing the results, with binder stiffness and bond 109 AH GREYLING

December 2012


strength increasing with time. DSR strain sweep results suggest that the BBS test is capturing a fundamental engineering property in bond strength. The BBS-DSR correlation reinforces the notion that bond strength development can be quantified as the emulsion breaks and begins to release water.

Figure 9-14- BBS test results compare well to DSR strain sweep results at two strain levels.

9.8 INTER-LABORATORY EVALUATION OF BBS TEST METHOD Based on the results achieved at the University of Stellenbosch and at UWM it was possible to validate the BBS Test method.

The South African evaluation utilized an identical experimental setup (e.g. PQG testing

instrument and modified pull-out stubs) for a different set of emulsified binders and substrate types.

Substrate preparation procedures differed slightly due to differences in available substrate preparation equipment. UWM tests utilized large aggregate plates that accommodated four pull-out stubs, while US tests utilized aggregate slices from cored rock samples that accommodated only a single pull-out stub.

Figure 9-15 and Figure 9-16 depict results for different emulsion types and substrate types, respectively. These findings corroborate the findings of Miller and research personnel at UWM that 1) bond strength development can be quantified over time; 2) the BBS test method can characterize bond strength development for different emulsion types; and 3) that advantageous combinations of aggregate-binder can be identified using the BBS test method. 110 AH GREYLING

December 2012


Figure 9-15- Pull-out tension values differ for various emulsion types at a range of curing intervals.

Figure 9-16- Pull-out tension values differ for various substrate types at a range of curing intervals.

Table 9-11 displays experimental results from the inter-laboratory testing. Values of the coefficient of variation (COV) are typically less than 10 percent, indicating good repeatability for three replicates in spite of a different substrate preparation procedure.

111 AH GREYLING

December 2012


Table 9-11-BBS test results for materials tested at the University of Stellenbosch. Emulsion

TEST SETUP Aggregate Granite

CRS65 Tillite

Granite SS60 Tillite

Granite CRS65+ 3% LATEX Tillite

Granite SS60+ 3% LATEX Tillite

Curing Time 2 Hours 6 Hours 24 Hours 2 Hours 6 Hours 24 Hours 2 Hours 6 Hours 24 Hours 2 Hours 6 Hours 24 Hours 2 Hours 6 Hours 24 Hours 2 Hours 6 Hours 24 Hours 2 Hours 6 Hours 24 Hours 2 Hours 6 Hours 24 Hours

Number 16 15 15 16 15 15 7 6 7 7 6 7 8 6 6 8 6 6 7 3 2 7 3 3

Tensile Strength (kPa) Average St Dev St Error COV 424.62 4.69 2.71 1.11% 533.02 19.54 11.28 3.67% 919.66 46.65 26.93 5.07% 406.55 15.88 9.17 3.91% 765.19 77.24 44.60 10.09% 835.65 57.13 32.98 6.84% 482.44 13.37 7.72 2.77% 466.18 5.64 3.26 1.21% 545.67 106.95 61.75 19.60% 445.40 8.28 4.78 1.86% 459.85 9.77 5.64 2.12% 568.26 51.49 29.73 9.06% 439.08 4.14 2.39 0.94% 824.81 13.91 8.03 1.69% 989.22 169.88 98.08 17.17% 446.30 12.42 7.17 2.78% 846.49 60.44 34.89 7.14% 1095.81 104.83 60.52 9.57% 472.50 6.82 3.94 1.44% 505.92 12.42 7.17 2.45% 477.92 5.64 3.26 1.18% 449.92 6.82 3.94 1.52% 484.24 2.71 1.56 0.56% 717.31 52.76 30.46 7.36%

112 AH GREYLING

December 2012


10

CONCLUSION & RECOMMENDATIONS

10.1 CONCLUSION The aim of this study was to fulfil the six study objectives defined in Chapter 1 of this report. The literature review clearly identified the factors that significantly affect the bond strength development between bituminous binders and aggregates. Through the literature review as well as through the requirements developed by the move towards performance specifications it was possible to confirm the need for a test method that will be able to quantify and characterize bond strength development, adhesive properties and aggregate-binder compatibility.

By using the Bitumen Bond Strength test it was possible to evaluate the practicality and repeatability of the developed procedure and the results measured was successfully used to evaluate the bond strength development of modified and unmodified bitumen emulsions on tillite and granite aggregates. It was also possible to correlate the results achieved at the University of Stellenbosch with results from the University of Wisconsin-Madison due to the fact that testing took place at both institutions.

The development of the test and the inter-laboratory test results in essence reinforced the hypothesis that the BBS test protocol can be used to effectively evaluate bond strength of different emulsion types and aggregate types. Except for the loading rate which is a known critical influence, the emulsion type and curing intervals are both identified as the most significant other factors contributing to bond strength development. All these factors were confirmed as part of the test completed during this study.

Aggregate type is also identified as a significant factor that will influence the bond strength development. Interactions between emulsion type and curing interval are identified as the most significant interaction. Although this study delivered a good set of initial results, a lot of further validation test on the BBS test method is still required for the test to be integrated into a performance-based specification system for surface seals.

113 AH GREYLING

December 2012


10.2 RECOMMENDATION FOR FUTURE STUDIES

The development of the Bitumen Bond Strength Test has supplied the opportunity to conduct tests and research in a wide new field. The following areas should be investigated in more detail to answer some of the question raised during this study: 

Test should be conducted using a broader range of emulsions and aggregates. The results yielded will ultimately add to the testing database and well as ensure that the method is developed. .



Test regimes that include bitumen in its modified and unmodified state should be completed. This information can be used for the evaluation of bond strength development in regular bitumen chip seal.



Further tests should also be done to test the influences of curing times longer than 24 hours. This may show more insight into the nature of the specific breaking action and adhesion development relationship between different aggregates and emulsions.



Further studies should also look to address a wider range of curing temperatures and humidity to investigate the tensile strength sensitivity to these factors. The influence of subzero temperatures in the first 48 hours may deliver some interesting results.



There should be more studies done to see if the effects of pre coating fluids on the bitumen and aggregate bond can be quantified.



More correlation test should be done to confirm the proposed critical pull-off value of 850 kPa or any more suitable variations on this value.

_

114 AH GREYLING

December 2012


11

BIBLIOGRAPHY

A Epps, C. G. R. B., 2001. A Performance-Graded Binder Specification for Surface Treatments. s.l.:Texas Department of Transportation. AASHTO TP-91 , 2011. AASHTO TP-91- Standard Method of Test for Determining Asphalt Binder Bond Strength by Means of the Asphalt Bond Strength (ABS), s.l.: AASHTO. Asphalt Academy-TG1, 2007. Technical Guideline 1 : The use of Modified Bituminous Binders in Road Construction, Pretoria: Asphalt Academy. Asphalt Academy-TG2, 2009. Technical Guideline 2- Bitumen Stabilised Materials, A Guideline for the Design and Construction of Bitumen Emulsions and Foamed Bitumen Stabilised Materials, Pretoria: Asphalt Academy. ASTM D2397 - 05, 2012. ASTM D2397 - 05-Standard Specification for Cationic Emulsified Asphalt, USA: ASTM. ASTM D4541-09, 2009. ASTM D4541-09-Standard Test Method for Pull-Off Strength of Coatings Using Portable Adhesion Testers, West Conshohocken: American Society of Testing and Materials. ASTM D977-12, 2012. ASTM D977-12-Standard Specification for Emulsified Asphalt, USA: ASTM. Atanackovic, T. M. & Guran, A., 2011. Theory of elasticity for scientists and engineers. Boston: Birkhäuser. Austroads, 2004. Spray Seal Design Method, Australia: Austrpads. Bickford, A., 2001. Basic Emulsion Chemistry - AEMA Asphalt Emulsion Technoloogies Wokshop, Presentation. s.l.:Asphalt Innovations. Boussad, N. & Martin, T., 1996. Emulsifier Content in Water Phase and Particle Size Distribution: Two Key Parameters for teh Management of bituminous Emulsion Performance. s.l., s.n. BP Bitumen, 2011. BP Bitumen. [Online] Available at: http://www.bp.com/sectiongenericarticle.do?categoryId=9027521&contentId=7050109 [Accessed 03 20 2011]. Canestrari, F. et al., 2010. Adhesive and Cohesive Properties of Asphalt-Aggregate Systems Subjected to Moisture Damage. Journal of Road Materials and Pavement Design(11-32,2010). Constable, B. T. L., 2009. V10- Final Year Thesis - The evaluation of the Bitumen Bond Strength Test, Stellenbosch: s.n. CSIR-TRH6, 1985. TRH6-Nomenclature and methods for Describing the Condition of Asphalt Pavements, Pretoria: National Institure of Transport and Road Research. Curtis, C., Ensley, E. & Epps, J., 1993. Fundamental properties of asphalt-aggregate interactions including adhesion and adsorption, Strategic Highway Research Program, Report SHRP-A-341, Washington DC: National Research Council . Ecopave Australia, 2009. The History of Bitumen Asphalt Concrete Roads. [Online] Available at: http://www.ecopave.com.au/bio_bitumen_asphalt_concrete_research_ecopave_australia_013.htm Epps, J., Chaffin, C. & Hill, A., 1980. Field Evaluation of a Chip seal Design Method, Research Report 214-23, s.l.: Texas Transportation Institute.

115 AH GREYLING

December 2012


Epps, J., Galloway, B. & Brown, M., 1974. Synthetic Aggregate Chip seal Research Report 83-2F, s.l.: Texas Transportation Institute. Eurobitume, 2012. Eurobitume. [Online] [Accessed 2012]. European Commitee for Standarization, 1999. Bitumen and bituminous binders, Determination of cohesion of binders for surface dressings, prEN 13588, Brussels: CEN. Forum of European National Highway Research Laboratories, 2004. Analysis of Available Data for Validation of Bitumen Tests, Report on Phase 1 of teh BiTVal Project, Brussels: Forum of European National Highway Research Laboratories. Gorman, J., Crawford, R. & Harding , I., 2004. Bitumen Emulsions in road construction - A Review. Issue March 2004. Gransburg, D., Zaman, M. & Aktes, B., 2010. Analysis of Aggregates and Binders used for the ODOT Chip Seal Program, Oklahoma City: Oklahoma Department of Transportation. H Bahia, J. M. R. V. T. M. C. D. R. M., 2010. Evaluation of the Bitumen Bond Strength (BBS) Test for Moisture Damage Charaterization. Madision: University of Wisconsin. Hanson, F., 1935. The bitumitominous surface treatment of rurual highways. s.l., s.n., pp. 89-178. Jaixing Mide Machinery Company, 2012. Jaixing Mide Machinery Company. [Online] Available at: http://www.cnmide.com/english/down.asp James, A., 2002. Asphalt Emulsion ( Chemistry and Concepts). s.l.:Akzo Nobel Chemicals. James, A., 2006. Overview of Asphalt Emulsion. Transport Research Circulary Number E-C102, Asphalt Emulsion Technology, pp. 1-15. Kanitpong, K. & Bahia, H., 2003. Role of Adhesion and Thin Film Tackiness of Asphalt Binders in Moisture Damage of HMA. Asphalt Paving Technology, Volume 72, pp. 502-528. Kearby, J., (1953). Tests and Theories on Penetration Surfaces. Highway Research Board, Volume 32. Louw, K., Spence, K. & Kuun, P., 2004. The use of bitumen emulsion as a cost effective solution for constructing seals during winter.. Proceedings of the 8th Conference on Asphalt Pavements for Southern Africa (CAPSA'04), s.n. McLeod, N., 1969. Basic principles for the design and construction of seal coats and surface treatments, Minnesota, USA.: Association of asphalt paving technologists. Meng, J., 2010. Affinity of Asphalt to Mineral Aggregate : Pull-Off Test Evaluation - Master of Science Thesis, Madison: University of Wisconsin. Miller, T., 2010. Development of Bond Strength Test for Improved Characterization of Asphalt Emulsion- Master of Science Thesis, Madison: University of Wisconsin-Madison. Miller, T., 2010. Development of Bond Strength Test for Improved Characterization of Asphalt Emulsion- Master of Science Thesis, Madison: University of Wisconsin-Madison. Milne, T. I., 2004. Towards a performance related seal design method for bitumen and modified road seal binders, Dissertation presented for the Degree of Doctor of Philosophy, Stellenbosch: University of Stellenbosch. 116 AH GREYLING

December 2012


National Cooperative Highway Research Program, 2005. NCHRP Synthesis 342-Chip Seal Best Practices, Washington, DC 20001: National Cooperative Highway Research Program. Opus International Consultants, 2010. Performance based bitumen specification, discussion paper, New Zealand: s.n. Opus International Consultants, 2010. Performance Based Bitumen Specifications. New Zealand: s.n. Oxford University Press, 2011. Oxford Dictionaries. [Online] Available at: http://oxforddictionaries.com/ [Accessed 24 September 2011]. Rabiot, D. & B. G. Koenders, a. M. A., 1999. A Quality Concept for Emulsion Grade Bitumens.. Washington, s.n. Roberts, C. & Nicolls, J., 2008. Road Note 39-Design guide for road surface dresssing, Berkshire: IHS for TRL. Rodríguez-Valverde, M. Á., Cabrerizo-Vílchez, M. Á., Páez-Dueñas, A. & Hidalgo-Álvarez., R., 2002. Kinetic Model of Bitumen Emulsion Breaking. Lyon, France,, 3rd World Congress on Emulsions. SABITA, 2009. Research and reminiscences from roadmakers- New Graham Ross publication paves the way for future historic research. asphaltNEWS, 23(3). SABITA-M2, 2007. Manual 2 - Bituminous binders for road Bituminous binders for road, Howard Place: SABITA. SANS 309, 2004. SANS 309:2004-Anionic bitumen road emulsions, Pretoria: South African Buro of Standards. SANS 548, 2003. SANS 548:2003- Cationic bitumen road emulsions, Pretoria: SABS. SANS4001-BT1, 2012. SANS4001-BT1,Civil Engineering Test Methods - Part BT1 - Penetration Grade Bitumen, Pretoria: South African Buro of Standards. SANS4001-BT2, 2012. SANS4001-BT2 , Civil Engineering Test Methods - Part BT2 - Cutback Bitumens, Pretoria: South African Buro of Standards. Shell Bitumen, 2003. The Shell Bitumen Handbook. 5th ed. London: Shell Bitumen. South African National Road Agency, 2007. Technical Recommendations for Highways 3 (TRH3-2007) - Design and Construction of Surfacing Seals, Pretoria: South African National Road Agency. Sponholtz, S., 2012. A Brief History of Road Building. [Online] Available at: http://www.triplenine.org/articles/roadbuilding.asp Superpave, 2012. Pavement Interactive. [Online] Available at: http://www.pavementinteractive.org/article/superpave-performance-grading/ Texas Department of Transportation, May 2010. Seal Coat and Surface Treatment Manual, s.l.: Texas Department of Transportation. Towler, J. & Dawson, J., 2008. History of Chipsealing in New Zealand - Hanson to P/17. Adelaide, ARRB Group Ltd and Authors 2008. Verlhac, P., Verzaro, F., Calderon., F. L. & J. J. Potti, a. B. E., 2002. Characterisation of Bituminous Emulsions: Particle Size Distribution and Amount of Residual Emulsifier.. Lyon, France, s.n. Walter, J. a. D. D., 2002. Coalescence of Quick Set Surface Dressing PMB Emulsions. Lyon, France,, 3rd World Congress on Emulsions,. Wates, J. a. A. J., 1993. Zeta Potential Measurements of Bitumen Emulsions and Road Aggregates.. Paris, s.n. 117 AH GREYLING

December 2012


Weinert, H., 1980. The Natural Road Construction Materials of South Africa. Pretoria: Academia. Wikipedia - Rock Geology, 1996. Petrology. [Online] [Accessed Chisholm, Hugh, ed. (1911). "Petrology". Encyclopædia Britannica (11th ed.). Cambridge University Press.]. Youtcheff, J. & Aurilio, V., 1999. Moisture sentivity testing of bitumen using a pneumatic adhesion test, Luxembourg: Eurobitume Workshop.

118 AH GREYLING

December 2012


ADDENDUM 1- FINAL DRAFT OF BBS TEST PROCEDURE - LATER ACCEPTED AS (AASHTO TP-91 , 2011)

119 AH GREYLING

December 2012

Standard Method of Test for

Determining Asphalt Binder Bond Strength by Means of the Bitumen Bond Strength (BBS) Test AASHTO Designation: 1.

SCOPE

1.1

This test method quantifies the tensile force needed to remove a pullout stub from an asphalt binder adhered on a solid substrate. Samples are conditioned at a controlled temperature and humidity based on experimental conditions determined at the beginning of the investigation. A pneumatic load is applied to the pullout stub until failure using an ASTM Type IV adhesion tester. The pullout tension at failure is used to describe the adhesive properties of asphalt binders and the compatibility between particular aggregates and binders.

1.2

This test may involve hazardous materials, operations, and equipment. This test method does not purport to address all of the safety concerns associated with its use. It is the responsibility of the user to establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use.

2.

REFERENCED DOCUMENTS

2.1

ASTM Standards:

D 140, Practice for Sampling Bituminous Materials

D 977, Specification for Emulsified Asphalt

D 2397, Specification for Cationic Emulsified Asphalt

D 4541, Pull-Off Strength of Coatings Using Portable Adhesion Testers

E 77, Inspection and Verification of Thermometers

E 145, Gravity-Convection and Forced-Ventilation Ovens

3.

TERMINOLOGY

3.1

Definitions:

3.1.1

adhesion—bond strength between the substrate and asphalt binder.

3.1.2

cohesion—bond strength within the asphalt binder.

VERSION UPDATED BY TDM 03/22/2010

AASHTO Draft Standard

3.1.3

Saturated surface dry (SSD) condition—in the SSD condition, the aggregate’s voids are filled with moisture but the main surface area of the aggregate particles is dry.

4.

SUMMARY OF METHOD

4.1

The adhesion tester applies a pneumatic load via a pressure ring to a pullout stub fixed to the substrate with asphalt binder, as shown in Figure 1. The binder is adhered to the substrate and can be subjected to conditioning at fixed levels of temperature and humidity prior to testing. Moisture conditioning is also possible for hot-applied binders to evaluate the effects of moisture damage. Pullout stubs are allowed to acclimate to laboratory conditions prior to testing. The stress applied to the assembly is recorded over time, allowing for calculation of load to failure and loading rate. The surface of the substrate is visually examined to determine the type of failure mode.

Scale is approximate. Figure 1 – Schematic Representation of the Testing Assembly for the Bitumen Bond Strength Test

5.

SIGNIFICANCE AND USE

5.1

Pullout tensile strength values measured over a range of environmental conditions and curing times provide information related to the adhesive and cohesive behavior of hot and emulsified asphalt binders.

5.2

Evaluation of pullout tensile strength on different aggregate substrates can be used to evaluate asphalt-aggregate compatibility.

5.2.1

For emulsified binders, this relationship allows for comparison of materials based on curing rate and ultimate tensile strength.



5.2.2

For hot binders, the pullout tensile strength may be used to evaluate moisture damage as measured by the decrease in tensile strength due to moisture conditioning for a given asphalt-aggregate combination.

6.

APPARATUS

6.1

Molds—For emulsified binders, use a silicone mold measuring approximately 40 mm x 40 mm with a 20 mm-diameter hole and 0.8 mm thickness. The mold has no backing and is used to contain the emulsion on the aggregate surface. For hot binders, use a silicone mold measuring approximately 40 mm x 40 mm with a 10 mm-diameter cavity. This mold is similar to the molds used to prepare 8-mm Dynamic Shear Rheometer (DSR) test samples. Figure 2 depicts diagrams of each mold type.

Figure 2 – Mold Dimensions (mm) for Emulsified Binders (left) and Hot Binders (right) 6.2

Base Plate—A solid aggregate substrate, aggregate composite, or glass plate of sufficient thickness is used to maintain structural integrity during testing. For emulsified binders, the plate must be uniformly flat to ensure that the binder will not flow beneath the mold. For both binder types, the plate must be uniformly flat to reduce the possibility of eccentric loading when the pullout stub is tested.

6.3

Testing Machine—Use a Type IV adhesion tester as defined by ASTM D 4541 for all tests. Such a device must consist of, at minimum, a control module, pressure ring, pressure plate, and data capture software. While different loading fixtures are available, the F-2 size pressure ring and pressure plate work sufficiently well for both types of asphalt binders. Figure 3 depicts an example testing machine.



Rate Control

Test Button Reaction Plate Loading Direction

Piston Pullout Stub Piston Support Test Sample

Substrate

Figure 3 – General Representation of Bitumen Bond Strength Test Apparatus 6.4

Air Supply—Capable of producing a consistent air pressure of at least 100 psi (0.67 MPa) as read on the supply gauge. Self-contained air cylinders, shop (bottled) air, or air from an automatic pump may be used.

6.5

Pullout Stubs—Stainless steel or any other suitable material, with dimensions as shown in Figure 4. Stub edges are beveled to reduce the amount of binder trapped between the stub edge and substrate and to ensure a uniform film thickness.

40 7

15

6

22 20

5

0.8

Top View

Bottom View

Figure 4 – Pullout Stub Dimensions (mm) for Bitumen Bond Strength Test 6.6

Forced Draft Oven—Capable of maintaining temperatures of at least 150 +/- 3°C for preparing all aggregate and binder samples. Two temperature-controlled ovens of Type IIA or IIB as defined in ASTM E 145 should be used due to different heating conditions required in the sample preparation process.



6.7

Environmental Chamber—Capable of maintaining temperatures between 15°C and 75°C +/- 1°C, and relative humidity between 20 percent and 80 percent +/- 1 percent, for curing all emulsion samples.

6.8

Thermometer—For tests performed at 25°C (77°F), use ASTM Thermometer No. 17C (17F) to measure the temperature of the aggregate surface prior to testing. For tests performed at other temperatures, use ASTM thermometers of an appropriate range and accuracy equal to that of the No. 17 thermometer. Since the accuracy of the test results is dependent upon closely controlled temperature conditions, thermometers should be calibrated in accordance with ASTM E 77. Thermometric devices with the same accuracy as ASTM thermometers may also be used.

6.9

Container—Any suitable container may be used to hold the hot asphalt material while being melted. For emulsified binders, the suitable container may be plastic, non-metal, or epoxy-lined, if metal.

6.10

Ultrasonic Cleaner—An ultrasonic cleaner is needed to remove residual particles from substrate surfaces prior to testing. The ultrasonic cleaner should be capable of maintaining temperatures of 60°C and should have a chamber large enough to allow for complete submersion of the substrate.

7.

SAFETY PRECAUTIONS

7.1

Observe standard laboratory safety precautions when preparing and testing hot binders and emulsified binders.

8.

CALIBRATION OF TESTING EQUIPMENT

8.1

Check the data acquisition capabilities of the testing system to ensure that activation of the pressure ring produces a real-time transient plot of pullout tension.

8.2

Verify the operating condition of all physical components in the testing system (i.e., air supply, pressure ring, software, and connections) prior to testing.

8.3

Calibrate the testing system prior to initial use and as often as necessary per the manufacturer’s recommendation and instructions.

9.

AGGREGATE TEST SPECIMENS

9.1

Solid Aggregate Substrates—Cut aggregate substrates from either quarried rocks or cored rock samples using standard rock saws such that plate faces are parallel. Lap all substrates using a 280-grit silicon carbide material on a standard lapidary wheel to remove saw marks and to ensure a consistent surface roughness. Once cut, clean samples for 60 minutes in an ultrasonic cleaner containing distilled water at a temperature of 60°C to remove residual particles on the plate surface.

9.2

Composite Substrates—Composite substrates may be prepared by cutting slices of samples cast in standard Portland cement concrete cylinder molds. Composite substrates



contain aggregate chips and a rapid-setting cement compound. These substrates need not be lapped but may be cleaned in the ultrasonic cleaner to remove dust and residual particles.

10.

HOT BINDERS

10.1

Sample Preparation:

10.1.1

Obtain a representative sample of the material for testing using procedures specified in ASTM D 140.

10.1.2

Handle everything with clean laboratory gloves. Place pullout stubs and substrates in a forced-draft oven at 60 ± 2°C for a minimum of 30 minutes to remove any residual moisture from the substrate surface and to prepare pullout stubs for application.

10.1.3

Simultaneously heat the hot binder in an appropriate container to 150 ± 2°C in a second forced-draft oven.

10.1.4

Pour molten binder into mold cavities. Air cool for 15 minutes and then trim samples to a weight of 0.4 ± 0.05 g. Air cool samples for an additional 20 minutes. If samples are not easily removed from the molds, refrigerate for an additional 10 to 15 minutes before removing.

10.1.5

Remove pullout stubs from the oven and place demolded binder samples onto the stub surface. Within 1 to 2 minutes, place hot pullout stub and binder sample on the substrate surface. Binder samples should be positioned on the substrate to allow for placement of multiple samples and for sufficient clearance of the testing apparatus.

10.1.6

Firmly press pullout stub down on substrate surface to ensure good adhesion between binder, stub, and substrate. Expect excess binder to flow out of the pullout stub channels. Avoid twisting the pullout stub to reduce air entrapment between the binder sample and substrate surface.

10.1.7

Allow dry samples to acclimate to lab conditions for approximately 24 hours before testing.

10.1.8

For wet-conditioned samples, allow samples to acclimate to lab conditions for 1 hour before wet conditioning. Place samples in wet (submerged) environments at 40°C ± 2°C for a predetermined conditioning interval. Water or de-icing agents may be used for wet conditioning.

10.1.9

Allow conditioned samples to acclimate to lab conditions for 1 hour before testing.

10.2

Test Procedure:

10.2.1

Record temperature of the substrate before testing samples.

10.2.2

Place four spacers on opposite sides of the sample to be tested to reduce eccentric loading and to allow the stub to fully separate from the substrate surface.



10.2.3

Gently place pressure ring around the pullout stub and resting on the spacers so as not to disturb the stub or induce unnecessary strain in the sample.

10.2.4

Gently screw pressure plate onto the pullout stub, taking care not to rotate the stub, until the pressure plate just touches the gasket.

10.2.5

Unscrew the load plate one quarter of a turn (approximately 90°) to ensure a physical gap between the plate and the gasket.

10.2.6

Test the samples using the adhesion testing device in accordance with the manufacturer’s recommendations. Record laboratory conditions (temperature and humidity) for all observations.

10.2.7

Record the maximum pullout tension and observe the failure mode. If more than 50 percent of the substrate surface remains exposed following removal of the pullout stub, the failure is considered adhesive. If less than 50 percent of the substrate surface remains exposed following the removal of the pullout stub, the failure is considered cohesive.

10.2.8

Repeat test procedure until all samples have been tested. A minimum of three samples should be tested at each set of experimental conditions.

11.

EMULSIFIED BINDERS

11.1

Sample Preparation:

11.1.1

Obtain a representative sample of the material for testing using procedures specified in ASTM D 140.

11.1.2

Heat emulsified binder to an application temperature of 60 ± 2°C in a forced-draft oven. Heat emulsion in a covered plastic container or non-metallic container for no longer than 1.5 hours to avoid premature breaking.

11.1.3

Simultaneously heat aggregate substrates to an application temperature of 25 ± 2°C in a second forced-draft oven.

11.1.4

If moisture conditioning samples, allow the substrates to soak in a bath of heated distilled water to achieve the saturated surface dry (SSD) condition while preheating.

11.1.5

Place molds without backing on the substrate surface and fill with liquid emulsion using a plastic eyedropper. Binder samples should be positioned on the substrate to allow for placement of multiple samples and for sufficient clearance of the testing apparatus.

11.1.6

Cure the substrate and filled molds under controlled conditions in an environmental chamber for a given curing interval.

11.1.7

While samples are curing, heat pullout stubs to 60 ± 2°C in a forced-draft oven.

11.1.8

After removing samples from the environmental chamber, remove the silicone molds encircling the cured binder and place the heated pullout stubs on the binder samples.



11.1.9

Firmly press pullout stub down on substrate surface to ensure good adhesion between binder, stub, and substrate. Expect excess binder to flow out of the pullout stub channels. Avoid twisting the pullout stub to reduce air entrapment between the binder sample and substrate surface.

11.1.10

Return the testing assembly to a forced draft oven at 25°C ± 2°C for approximately 1 hour to allow the samples to acclimate to testing conditions.

11.2

Test Procedure:

11.2.1


11.2.2

Place four spacers on opposite sides of the sample to be tested to reduce eccentric loading and to allow the stub to fully separate from the substrate surface.

11.2.3

Gently place pressure ring around pullout stub and resting on the spacers so as not to disturb the stub or induce unnecessary strain in the sample.

11.2.4

Gently screw pressure plate onto the pullout stub, taking care not to rotate the stub, until the pressure plate just touches the gasket.

11.2.5


11.2.6


11.2.7

Record the maximum pullout tension and observe the failure mode. If more than 50 percent of the substrate surface remains exposed following removal of the pullout stub, the failure is considered adhesive. If less than 50 percent of the substrate surface remains exposed following the removal of the pullout stub, the failure is considered cohesive.

11.2.8

Repeat test procedure until all samples have been tested. A minimum of three samples should be tested at each set of experimental conditions.

12.

REPORT

12.1

Obtain the pullout tension directly from the graphical computer interface and report to the nearest 0.1 psi (0.7 kPa). Environmental conditions (i.e., temperature) should be noted for each test.

12.2

Pullout tension results should be averaged for each set of samples tested. A minimum of three samples should be tested at each set of experimental conditions.

12.3

Determine the loading rate by calculating the slope of the line between the initial pullout tension and the final failure tension. The slope of the line should be calculated between 20 percent and 80 percent of the maximum stress.



12.4

Document the failure mode as cohesive, adhesive, or a combination of failure modes through visual observation and photo analysis. Reject samples that exhibit adhesive failure between the pullout stub surface and asphalt binder due to inadequate stub coverage.

13.

PRECISION AND BIAS

13.1

Pullout tension results should exhibit a coefficient of variation less than or equal to 10 percent to be considered valid.




ADDENDUM 2- UWM STUDY TOUR FEEDBACK REPORT

120 AH GREYLING

December 2012

________________________________ UNIVERSITEIT VAN STELLENBOSCH UNIVERSITY OF STELLENBOSCH

SUMMARY REPORT

25 AUGUST 2008– 04 SEPTEMBER 2008

STUDY VISIT TO THE UNIVERSITY OF WISCONSIN MADISON

PREPARED FOR:

PREPARED BY:

PROF KIM JENKINS PROF HUSSAIN BAHIA

André Greyling US# 12535605 [email protected] Cell No: +27 83 962 3040 Tel No: +27 21 527 7000 Fax No: +27 21 527 7001

VISIT TO UW-MADISON- SUMMARY REPORT

TABLE OF CONTENT

1

INTRODUCTION ............................................................................................................................. 3 1.1 BACKGROUND .......................................................................................................................... 3 1.2 RESEARCH TOPIC ..................................................................................................................... 3 2 ADHESION TEST ............................................................................................................................. 4 2.1 TEST DETAILS AND PROCEDURE ........................................................................................ 4 2.1.1 2.1.2 2.1.3 2.1.4

2.2 2.2.1 2.2.2 2.2.3 2.2.4 2.2.5

3

EMULSION SAMPLES ............................................................................................................................ 4 ADHESION MEDIUM ............................................................................................................................. 5 TESTING PROCEDURE .......................................................................................................................... 5 TEST METHODS....................................................................................................................................... 7

TEST RESULTS ............................................................................................................................ 9 6 HOURS .................................................................................................................................................... 9 24 HOURS ................................................................................................................................................ 11 30 HOURS ................................................................................................................................................ 12 48 HOURS ................................................................................................................................................ 14 COMBINED RESULTS ........................................................................................................................... 16

CONCLUSION & DISCUSSION ................................................................................................. 18 3.1 CONCERNS AND OPEN QUESTIONS ................................................................................ 18 3.2 PROCEDURES FOR DISCUSSION ........................................................................................ 19 3.3 CONCLUSION .......................................................................................................................... 19

Prepared by AH GREYLING US# 12535605

2


1 1.1

INTRODUCTION BACKGROUND A study visit was conducted by Professor Kim Jenkins and Mr Andre Greyling from the University of Stellenbosch to the University of Wisconsin Madison during August and September 2008. The main purpose of the visit was to identify an area of study in which research done by the University of Stellenbosch could assist and be combined with research done by University of Wisconsin Madison. This report serves as a summary of the observations made, lessons learnt and matters of interest identified as an outcome of this visit and it also serves as a starting point for various discussions to follow.

1.2

RESEARCH TOPIC During various informal discussions over possible research topics, Asphalt Emulsion Adhesion/Cohesion properties were identified as a suitable area of research. Further discussions were held on the possible factors influencing Asphalt Emulsion Adhesion/Cohesion properties and the following areas of relevance were identified. These various points will be discussed in greater details in future reports. The Factors Affecting the Asphalt Emulsion- Aggregate Bond Strength Influencing Factor Possible Options Additional Options Responsible 1

Surface Roughness

2

Temperature

3

Bitumen Source

4

Aggregate Source

5

Time Of Curing

6

Moisture- Stone

7 8

Rate of Loading Film Thickness

9

Relevant Test method

Smooth Rough Stone Bitumen Chemistry Viscosity -Rheology Type

Recovery Method Chemistry Surface Pores 6 Hours 24 36 48 Wet Dry 0.4 mm Posi Patti

Emulsion Foam Reference Binder

US US UW UW US US US US US US US US US US US US US US UW UW UW UW

Factors 2

5

4

4

2 2 1 2


3


2

ADHESION TEST On identification of the relevance and importance of the various factors mentioned above, the opportunity was used to investigate and test the relevance of the current two testing methods mentioned under Point 9 above. The UWM has previously done test using both the Patti and the Posi test methods. A test was therefore formulated to investigate which of the two tests would be the best and most relevant to use for the proposed research. (It must be noted that the results should not be seen as accurate but only as indicative for the purposed of evaluating the relevance of the test methods.)

2.1

TEST DETAILS AND PROCEDURE An outline of the formulated test can be summarized as follows: Two bitumen emulsion samples where tested, a CRS-2 and a Latex Modified CRS-2 emulsion. Both emulsions were recovered using the proposed standard method described in the “Standard Practice for Recovering Residue Using an Evaporative Technique for Emulsified Asphalts”. Samples of the recovered binder were taken at 6, 24, 30 and 48 hours. These samples were placed in silicon moulds and allowed to cool down to room temperature. The Posi and Patti aluminium stubs as well as glass plates were heated to 60°C. The binders in the moulds were first applied to the glass plates after which the aluminium stubs were applied. The stub spacers/supports were used to ensure a uniform film thickness of 0.4mm. The prepared samples were left to cure for two hours after which they were tested using the Patti and Posi Tests. A further more detailed description and the results of the tests follow below:

2.1.1

EMULSION SAMPLES The following two emulsion samples were tested:  CRS-2 is cationic rapid setting type emulsified 120-150pen asphalt. This emulsion is used primarily for the construction of bituminous surface treatments (chip seals).  LMCRS-2 is a high performance cationic rapid set emulsion made with a standard 120150pen binder modified with polymer latex. LMCRS-2 are specifically formulated for high performance chip seals. The polymer creates higher initial strength, higher stone retention over the life of the seal and enhanced cracking resistance. Thermal susceptibility is reduced making seals with a greater performance range from sub zero to elevated temperatures. The latex in the system provides a highly flexible binder of elevated softening point and elastic properties.1

1

http://www.emultech.com/index.htm?page=/asphaltemulsions_cati_ani.htm


4


2.1.2

ADHESION MEDIUM Glass plates were used as the base medium with the aluminium Posi and Patti pullout stubs as the second applied medium. The pictures below show the typical setup of the Patti pullout stub on the left hand side of the picture and the Posi on the right. Note the pullout stub support/spacer.

Photo 1 – Pullout Stubs 2.1.3

TESTING PROCEDURE The following steps were followed as part of testing procedure:

2.1.3.1

RECOVERING PROCESS The Proposed Standard Practice for Recovering Residue using an Evaporative Technique for Emulsified Asphalts was used. This entailed spreading the two types of Emulsified asphalt onto a silicon mat at a spreading rate of 2.0kg/m2. The emulsion was placed in a 25°C forced draft oven for 24 hours followed by 24 hours at 60°C. Samples were taken from the recovered specimens at 6, 24, 30 & 48 Hours.


5


2.1.3.2

BITUMEN SAMPLING 6& 24 HOURS After 6 and 24 hours a sample was taken from the recovery tray at 25°C. Due to the still relatively high water content at both intervals it was necessary to form the samples manually using gloved hands. The samples were placed into silicon moulds using a steel spatula at room temperature. The picture below shows the typical silicon moulds.

Photo 2 – Pullout Stubs and Silicon Moulds details 30&48 HOURS After 30 and 48 hours samples were taken from the recovery tray at 60°C. At this stage the bitumen became very adhesive and it was not practical to handle the bitumen by hand or unheated spatula. It was therefore necessary to heat the spatulas over an industrial heat gun to make handling of the samples possible. This led to the immediate evaporation of any moisture left in the binder sample. This is noted as a practical matter of concern. 2.1.3.3

TEST SPECIMEN PREPARATION The prepared specimens in the silicon moulds were left to cool down to room temperature. The mould was then turned over on the pre heated 60°C glass plate and pressed down manually. The mould was removed immediately and the 60°C aluminium pullout stubs was applied to the top of the sample using the support/spacer blocks to ensure an accurate film thickness. These samples were then left to cure at room temperature for two hours before being tested. Prepared by AH GREYLING US# 12535605

6


2.1.3.4

CLEANING OF TEST EQUIPMENT All equipment used during the test procedure was cleaned using ZEP Big Orange Industrial Degreaser. Before the application of any Bitumen the surfaces was re cleaned using Acetone.

2.1.4

TEST METHODS One of the purposes of this test was to evaluate the effectiveness and relevance of the two current test methods. The two methods are briefly described below:

2.1.4.1

PATI TEST The Patti test was originally developed by the painting/adhesive industry to measure the adhesion or pull-off strength of a coating on solid. This testing method measures the greatest perpendicular force that a solid surface-coating can take before the adhesive is detached from the solid surface. The test also allows for the evaluation of the type or failure: adhesive (at the coatingsolid interface) or cohesive (within the coating) by inspecting the failure surface after the detachment has occurred. 2

Certain modifications were made to the PATTI test setup and methodology. The test was modified to enhance the parameters measured during the pullout test and to allow an improved evaluation of the mineral-binder adhesive and binder cohesive properties. The details of the modification are discussed in a report by Fratta & Daranga3. The following is the typical step that needs to be followed to complete the Patti Test:  Switch the Patti machine on and ensure that the data card reader programme is started on the connected PC.  Ensure that the relevant inputs on the programme screen are completed correctly.  Ensure that the air pressure supply is open and functional.  Do a test run to ensure that the data card reader is capturing data.  Put the Patti pressure ring over the prepared samples pullout stub.  Screw the self aligning ring in gently until resistance is felt.  Unscrew the self aligning ring half a turn.  Press Record on the data card reader programme and press the test button on the Patti Console.  Save the test results. The schematic below shows the typical Patti Test setup. (In this specific case the LVDT was not used as part of the test.) 2“ Dante Fratta and Codrin Daranga May 2008,Experimental validation of the modified PATTI Test methodology “ 3 “Dante Fratta and Codrin Daranga May 2008,Experimental validation of the modified PATTI Test methodology” Prepared by AH GREYLING US# 12535605

7


5 0 kHz. 8 channel DAQ card

Three - leg LVDT frame LVDT

Self - aligning screw plate Gasket Pressure transducer

Gasket

Pressurized gas inlet

To PATTI test controller

Pull stab

Binder

Space (Pullout stub support)

Te st s urface

2.1.4.2

POSI TEST The Posi Test is a digital adhesion tester. It is a hydraulic pull-off tester that measures the adhesion (bond strength) of coatings to metal, wood, concrete and other substrates. The typical setup consists of 20mm aluminum pullout stubs, a Hydraulic Hand pump and a self aligning quick coupling piston. The Test has a digital recorder that captures all the relevant test data. The picture below shows more details on the Test Equipment. As with the Patti test, the Posi test was not specifically developed to test the bitumen adhesion and certain modifications was made to make testing of bitumen binders possible.

Picture 2 – Posi Test Accessories Prepared by AH GREYLING US# 12535605

8


The following is the typical steps that need to be followed to complete the Posi Test:      

Switch on the digital recorder , set the stub size to 20mm and press clear. Open the valve and press the piston sleeve all the way down manually. Ensure the piston sleeve is screwed in completely. Apply the piston to the prepared samples’ pullout stub using the quick coupling. Apply pressure slowly until sleeve touches the spacers blocks. Lift the handle and push down slowly while trying to maintain a constant rate of pressure applications.  Upload the data from the Posi Test to a PC. 2.2

TEST RESULTS The methodology and the two different test methods described were used to perform various tests. The Results of these tests at the different recovery times can be summarized as follows:

2.2.1.1

6 HOURS PATTI During this specific test only one of the four samples was tested successfully. The reason for this being that none or very little adhesion formed between the pullout stub and the glass surface after only 6 hours of recovery. The result of the one successful test is shown in the graph below: CRS-2 - 6 Hours 7400

CRS-2 - 6 Hours

6900 Pressure (kPa)

2.2.1

6400 5900 5400 4900 4400 3900 0

0.5

1

1.5

2

Time (s)

Graph 1 – CRS2- 6 Hours –Patti Test Results The following comments are relevant to the information in the graph above: Prepared by AH GREYLING US# 12535605

9


 Note the maximum pullout pressure of approximately 4600kPa.  The rate of pressure application ranges from 9,000-11,000 kPa/s as a constant setting.  The maximum pullout pressure was calculated using the Voltage to Pressure conversion equation.  Adhesion failure to the glass medium. 2.2.1.2

POSI Due to the larger Pullout Stub size of the Posi test it was possible to test all four samples successfully. The results are summarized in the table below: Test CRS-2-Test 1 CRS-2-Test 2 LMCRS-2-Test 1 LMCRS-2-Test 2

Pressure (psi) 154 143 169 140

Pressure (kPa) 1061.79 985.95 1165.21 965.27

Rate (psi/sec) 337.50 172.00 431.20 34.40

Rate (kPa/sec) 2326.98 1185.90 2973.02 237.18

Duration (sec) 0.16 0.25 0.16 1.16

DollySize (mm) 20 20 20 20

Table 1 – PosiTest results – 6 Hours The following comments are relevant to the information in the table above:  Note the varying rate of pressure application. This is due to the manual application of pressure.  Note the large difference in the Maximum Pullout pressure of the Posi vs Patti. The results of the Posi test are approximately a quarter of the value of Patti test.  Note the large difference in the rate of pressure application of the Posi vs Patti.


10


2.2.2 2.2.2.1

24 HOURS PATTI

24 Hours 7400 CRS-2-24 Hours Trail 1

Pressure (kPA)

6900

CRS-2-24 Hours-Trail 2

6400

LMCRS-2-24 Hours Trail 1

5900

LMCRS-2-24 Hours Trail 2

5400 4900 4400 3900 0

0.5

1

1.5

2

Time (s)

Graph 2 – 24 Hours –Patti Test Results The following comments are relevant to the information in the graph above:  Note the Maximum pullout pressure results are in the range between 5000kPa and 5500kPa.  The results for the four trails and the two types of Emulsion correlate well. CRS2-Trail 1 and LMCRS2-Trail 2 shows almost exactly the same curve.  The rate of pressure application ranges from 9,000-11,000 kPa/s as a constant setting 2.2.2.2

POSI

Test CRS-2-Test 1 CRS-2-Test 2 LMCRS-2-Test 1 LMCRS-2-Test 2

Pressure (psi) 135 237 135 2494

Pressure (kPa) 930.79 1634.06 930.79 17195.52

Rate (psi/sec) 53.00 856.20 60.30 78.90

Rate (kPa/sec) 365.42 5903.29 415.75 544.00

Duration (sec) 0.66 0.16 0.58 30.32


Table 2 – Posi Test results – 24 Hours The following comments are relevant to the information in the table above: Prepared by AH GREYLING US# 12535605

11


 Note the varying rate of pressure application. This is due to the manual application of pressure.  Note that the results of LMCRS-2 Test 2 are not valid.  Note the large difference in the Maximum Pullout pressure of the Posi vs Patti. The results of the Posi test is approximately a fifth of the value of Patti test .  Note the large difference in the rate of pressure application of the Posi vs Patti

2.2.3.1

30 HOURS PATTI During the first two test only the information captured by the data card reader was evaluated. This is the data as shown in the graphs. For the 30 & 48 hours tests it was decided to capture the data as displayed on the programme screen directly after completion of the test as well. This led to the observation that there is a discrepancy between the saved and the displayed data. The reason for this is unknown but may relate to the voltage to pressure conversion factors.

30 Hours 7400 CRS-2-30 Hours - Trail 1

Pressure (kPA)

2.2.3

6900

CRS-2-30 Hours -Trail 2

6400

LMCRS-2-30 Hours -Trail 1 LMCRS-2-30 Hours -Trail 2

5900 5400 4900 4400 3900 0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

Time (s)

Graph 3 – 30 Hours –Patti Test Results The following comments are relevant to the information in the graph above:  Note the Maximum pullout pressure results are in the range between 5950kPa and 7350kPa.


12


 The results for the four trails and the two types of Emulsion correlate well.  The rate of pressure application ranges from 9,000-11,000 kPa/s as a constant setting The manually captured data shows an unexplained discrepancy. The Pullout rate relates well to the slope of the curve. The Maximum pullout tension is however approximately 50% of value of the data from the curves. The data that was captured manually from the computer screen is shown below. The data from the graphs is shown in brackets CRS-2-Trail 1  Maximum pullout tension =3384.73kPa (7350kPa)  Pullout Rate =10047.82kPa/s  Cohesive failure CRS-2-Trail 2  Maximum pullout tension =3025.10kPa (7000kPa)  Pullout Rate =10391.49kPa/s  Cohesive failure LMCRS-2-Trail 1  Maximum pullout tension =2125.68kPa(6100kPa)  Pullout Rate =10358.56kPa/s  Adhesive failure LMCRS-2-Trail 2  Maximum pullout tension =1978.51kPa(5900kPa)  Pullout Rate =10103.68 kPa/s  Adhesive failure It is also interesting to note the different failure mechanisms between the Modified and Unmodified emulsions.


13


2.2.3.2

POSI Pressure (psi) 204 379 189 213


Pressure (kPa) 1406.53 2613.11 1303.11 1468.58

Rate (psi/sec) 1116.00 556.00 226.00

Rate (kPa/sec) 7694.55 3833.49 1558.22

Duration (sec) 0.25 0.16 0.50


Table 3 – PosiTest results – 30 Hours The following comments are relevant to the information in the table above:  Note the varying rate of pressure application. This is due to the manual application of pressure.  Note that the results of CRS-2 Test 1 are not valid.  Note the large difference in the Maximum Pullout pressure of the Posi vs Patti. The results of the Posi test is approximately a fifth of the value of Patti test.  Note the large difference in the rate of pressure application of the Posi vs Patti

2.2.4.1

48 HOURS PATTI 48 Hours 7400 CRS-2-48 Hours-Trail 1

6900

CRS-2-48 Hours - Trail 2

Pressure (kPa)

2.2.4

6400

LMCRS-2-48 Hours -Trail 1 LMCRS-2-48 Hours-Trail 2

5900 5400 4900 4400 3900 0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

Time (s)

Graph 4 – 48 Hours –Patti Test Results


14


The following comments are relevant to the information in the graph above:  Note the Maximum pullout pressure results are in the range between 6500kPa and 7300kPa.  The results for the four trails and the two types of Emulsion correlate well.  The rate of pressure application ranges from 9,000-11,000 kPa/s as a constant setting The data that was captured manually from the computer screen is shown below. The data from the graphs is shown in brackets CRS-2-Trail 1  Maximum pullout tension =3221.22kPa(7160kPa)  Pullout Rate =10160.21kPa/s  Cohesive Failure CRS-2-Trail 2  Maximum pullout tension =3123.11kPa(7080kPa)  Pullout Rate =10337.00 kPa/s  50/50 Cohesive Adhesive failure LMCRS-2-Trail 1  Maximum pullout tension =3106.76kPa(7060kPa)  Pullout Rate =10141.95kPa/s  Cohesive Failure LMCRS-2-Trail 2  Maximum pullout tension =2632.57kPa(6570kPa)  Pullout Rate =-613.418 kPa/s (Card reader fault)  Cohesive failure 2.2.4.2

POSI


Pressure (psi) 202 430 365 464

Pressure (kPa) 1392.74 2964.75 2516.59 3199.17

Rate (psi/sec) 638 660 1656 1456

Rate (kPa/sec) 4395.41 4550.54 11419.10 10038.77

Duration (sec) 0.160 0.500 0.160 0.250


Table 4– Pos iTest results – 48 Hours The following comments are relevant to the information in the table above:


15


 Note the varying rate of pressure application. This is due to the manual application of pressure.  Note the large difference in the Maximum Pullout pressure of the Posi vs Patti. The results of the Posi test is approximately half of the value of Patti test.  Note the large difference in the rate of pressure application of the Posi vs Patti

PATTI TEST The two graphs below show the combined results for the two types of emulsions from the Patti test. The results generally correlate well and even though only two samples were tested for each scenario. The order of the results shows definite repeatability in the test. CRS-2 COMBINED 7900 CRS-2-24 Hours Trail 1

7400

CRS-2-24 Hours-Trail 2

Pressure (kPA)

2.2.5.1

COMBINED RESULTS

6900


6400

CRS-2-30 Hours -Trail 2 CRS-2-48 Hours-Trail 1

5900


5400 4900 4400 3900 0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

1.4

1.6

1.8

2

Time (s)

Graph 5 – Combined CRS-2 Results LMCRS-2 COMBINED 7400 LMCRS-2-24 Hours Trail 1 6900

LMCRS-2-24 Hours-Trail 2 LMCRS-2-30 Hours - Trail 1

6400 Pressure (kPa)

2.2.5

LMCRS-2-30 Hours -Trail 2 LMCRS-2-48 Hours-Trail 1

5900

LMCRS-2-48 Hours - Trail 2 5400 4900 4400 3900 0

0.2

0.4

0.6

0.8

1

1.2

Time (s)

Graph 6 – Combined LMCRS-2 Results


16


The following comments are relevant to the graphs above  From the slope of the curve we can see that the rate of application is fairly constant for all the samples tested  The results of the trails for the various recovery times correlate fairly well and show signs of repeatability. POSI TEST COMBINED POSI TEST RESULTS

3500.00

MAX PULLOUT PRESSURE

2.2.5.2

CRS2- Trail Set 1

3000.00

CRS2-Trail Set 2 2500.00

LMCRS2-Trail Set 1

2000.00

LMCRS2-Trail Set 2

1500.00 1000.00 500.00 0.00 0

10

20

30

40

50

60

RECOVERY TIME (HOURS)

Graph 7 – Combined Posi Test Results The following comments are relevant to the graph above  The Posi Test results is generally about 20% of the value of the Patti Test Results  There is a high variability in the rate of pressure application.  The results for the 6 hour test correlates well but the other three test sets shows highly variable answers.


17


3

CONCLUSION & DISCUSSION Although the results and data captured had some discrepancies it is believed that the completed test served its purpose. This section aim to summarize the observations, concerns and open question that were identified during the testing process.

3.1

CONCERNS AND OPEN QUESTIONS The following concerns and questions were raised during testing process: (a)

(b)

(c)

(d)

(e)

(f) (g)

(h)

(i) (j)

The Emulsion binder recovery does not take place on an aggregate medium. What relevance does that have to predicting adhesion in practice? Should recovery not take place on an aggregate medium? The emulsion is applied to the glass medium and the pullout stub at room temperature. Although this is not a big concern it must be noted that room temperatures can vary significantly. The Emulsion samples supplied showed signs of ageing. There were lumps of bitumen in both samples. Will it not be wise to specify a standard mixing procedure for the emulsion samples before starting the recover process? The only way to handle and form the recovered binder after 6 and 24 hours was by hand. At this specific time the binder still has fairly high water content and it is difficult to shape. It raises a concern of repeatability and consistency. The binder was applied to the 60°C glass plates at room temperature and the pullout stubs was applied to the top of the binder heated at 60°C. This method again raises concern about the relevance to what happen in practice. For example, chip seals; the binder would be applied at 60-70°C and the aggregate at ambient temperature. The question was raised about the relevance of curing the prepared sample for 2h @ 25°C. What is the purpose of this ?This should be investigated further. A serious concern is the rate of application of pressure on the Posi Test. To repeat two tests at the same rate of pressure application is impossible as pressure is applied manually. As the adhesion and cohesion failure pressure is a function of the rate of application, it raises serious concerns about the relevance and repeatability of the test method. Another practical concern is the preparation of samples after 30&48 hours at 60°C. At this stage and temperature the binder is sufficiently tacky to make molding and handling it impossible. To make the preparations of the samples possible, a metal spatula was heated and used to mould the binder samples. A visual observation was made that steam (i.e. water) is released from the binder when it is shaped with the heated spatula. This technically alters the properties of the recovered binder. The binder recovery and sample preparation will need to be reinvestigated. The discrepancy in the displayed and stored data from the Patti Test should be investigated. The Patti test seems to be the most accurate and repeatable test. The only negative being the cost. In discussions with UWM it came to light that it may be possible to construct a cheaper version of the Patti Test for use at US. This must be discussed and investigated further.


18


3.2

PROCEDURES FOR DISCUSSION The need for new procedures on the following issues was identified during the testing process.

3.2.1.1

PREPARATION OF AGGREGATE SPECIMEN AND RECOVER OF BINDER The recovery, preparation of the sample and application of the binder to a glass medium is not relevant to adhesion/cohesion in practice. A procedure that is relevant to the construction process in practice needs to be developed. It should include details on the following topics:          

3.3

The type of aggregate. The size of the prepared aggregate sample. The volume and surface area of the emulsion that should be applied? The rate of application of emulsion. Special provisions that should be made to ensure testing of the sample with the Patti test would be possible? The prepared surface of the aggregate sample. The temperature at which the emulsion is applied to the aggregate. The temperature of the aggregate when the emulsion is applied. The recovery/curing regime for the emulsion. 24h @ 25°C and 24h @ 60°C ??? The Testing regime – 6,24,36 & 48 h ???

CONCLUSION The following are preliminary conclusions of the tests performed:  The Posi Test does not show repeatable results .It is mainly due to the fact that the rate of pressure application can not be controlled. This shortcoming does not make it practical to perform asphalt adhesion tests in its current form.  The Patti Test show generally repeatable results with the only concern being the discrepancy in the displayed and stored data.


19


ADDENDUM 3- MINUTES OF STUDY GROUP CONFERENCE CALLS AND REPORTS COMPLETED DURING THIS PHASE

121 AH GREYLING

December 2012

Asphalt Research Consortium Conference Call with University of Stellenbosch and UW-Madison Date:

Thursday, 10 December 2009, 8:30 – 9:30 AM

Participants:

UW: Hussain Bahia, Andrew Hanz, Petrina Johannes, Tim Miller US: Andre Greyling, Kim Jenkins

Next meeting: Thursday, 07 January 2010, time TBD

Discussion Points Results from the latest rounds of testing. Make a final decision on experimental factors (temperature, substrate preparation). o Curing temperature to be 35 C. o UW working to prepare PCC slices. Will run some trials on existing slices, then improve mix design. Discs will be imaged to determine ratio of aggregate to cement. Completion of Year 3 testing. Year 3 ends March 2010. Development of AASHTO draft standard. Updates from Stellenbosch All tests being conducted on cored slices. 20 discs prepared for 2 aggregate types. Updates from Madison Correlations between sweep test and BBS test. Need to complete BBS testing by March 2010 to close out Year 3. Refer to existing experimental matrix for emulsions still to be tested. Outstanding Action Items Continue testing CRS-2 unmodified emulsion. (TM, ZA, PJ) Check on sandblasting procedures in geology department. (TM) Check with Codrin, Aaron, and Carl for information on the sand-patch procedure. (UW) Find rollers for sweep test. (HB, AH). Review ASTM standards for BBS testing. (UW, US) Determine relationship between aggregate retention and pullout tension. (UW, US) Expand BBS test procedure to include a section on HMA and a section on emulsions. (TM, UW) Contact Terence Milne regarding inclusion in conference calls. (KJ) Send updated BBS standard to Stellenbosch. (TM) Establish contractor survey for acceptance requirements or procedures for chip seals. (US, UW) Investigate MTD (mean texture depth) as method to measure bleeding. (AG) Research method to quantify opening time for seals. (AG) Conduct He porosimeter testing to measure porosity of aggregates. (TM) New Action Items Investigate field validation test methods. (AG)

Review experimental design and refine to meet current testing capabilities. (TM, AG) o Include 2 operators for all tests. o Select appropriate sample size (6-8 tests for selected variables). Benchmark instrumentation with hot binders and glass plates first. (UW, US)

DATE

Monday, November 30, 2009

SUBJECT

Bitumen Bond Strength (BBS) Testing Summary

TO

Andre Greyling, University of Stellenbosch, RSA Prof. Kim Jenkins, University of Stellenbosch, RSA Prof. Hussain Bahia, UW MARC Andrew Hanz, UW MARC Petrina Johannes, UW MARC

FROM

Tim Miller, UW MARC

Table 1. Emulsion samples by type and availability. Emulsion Properties Chemistry Set Modification None Rapid Latex Polymer Cationic None Slow Latex Polymer None Rapid (High Float) Latex Polymer Anionic None Slow Latex Polymer Total

Stellenbosch Availability

UW Availability

XXX XX XX X X

X X X X

Suggested Emulsions Stellenbosch UW

X X X

X X X X

X X 7

X X X X X X X

X X

8

5

7

Table 2. Experimental variables and levels. Level Variable Factors

1

2

3

4

5

6

dolomite

glass

granite US

tillite

CRS-65+3% latex

SS-66

A

Substrate Type

B

Emulsion Type

CRS-2

C

Curing Interval

2

6

24

D

Curing Temperature

25

35

45

E

Lab

UW

US

granite UW limestone

CRS-2P CRS-65

1

48

UW BBS Data 300

250

Glass / CRS-2 / 35 Granite / CRS-2 / 35 Limestone / CRS-2 / 35 Dolomite / CRS-2 / 35 Glass / CRS-2 / 45 Granite / CRS-2 / 45 Limestone / CRS-2 / 45 Dolomite / CRS-2 / 45 Glass / CRS-2P / 35 Granite / CRS-2P / 35 Limestone / CRS-2P / 35 Dolomite / CRS-2P / 35 Glass / CRS-2P / 45 Granite / CRS-2P / 45 Limestone / CRS-2P / 45 Dolomite / CRS-2P / 45

200 Pullout Tension (psi) 150

100

50

0 2

6

24

48

Curing Time (hr)

Figure 1. UW BBS data. US BBS Data 300 Granite / CRS-65 / 25 Tillite / CRS-65 / 25 Granite / CRS-65+latex / 25 Tillite / CRS-65+latex / 25 Granite / SS-66 / 25 Tillite / SS-66 / 25

250


100

50

0 2

6

24

Curing Time (hr)

Figure 2. US BBS data.

2

UW + US BBS Data All Combinations 300 Glass / CRS-2 / 35 Granite / CRS-2 / 35 Limestone / CRS-2 / 35 Dolomite / CRS-2 / 35 Glass / CRS-2 / 45 Granite / CRS-2 / 45 Limestone / CRS-2 / 45 Dolomite / CRS-2 / 45 Glass / CRS-2P / 35 Granite / CRS-2P / 35 Limestone / CRS-2P / 35 Dolomite / CRS-2P / 35 Glass / CRS-2P / 45 Granite / CRS-2P / 45 Limestone / CRS-2P / 45 Dolomite / CRS-2P / 45 Granite / CRS-65 / 25 Tillite / CRS-65 / 25 Granite / CRS-65+latex / 25 Tillite / CRS-65+latex / 25

250


100

50

0 2

6

24

48

Curing Time (hr)

Figure 3. UW and US BBS data. UW + US BBS Data Effect of Substrate 300

GLASS GRANITE UW LIMESTONE DOLOMITE GRANITE US TILLITE

250


100

50

0 2

6

24

48

Curing Time (hr)

Figure 4. Effect of substrate on pullout tension.

3

UW + US BBS Data Effect of Curing Temperature 300

250


100 25 35 45

50

0 2

6

24

48

Curing Time (hr)

Figure 5. Effect of curing temperature on pullout tension. UW + US BBS Data Effect of Emulsion 300

CRS-2 CRS-2P CRS-65 CRS-65 + 3% latex SS-66

250


100

50

0 2

6

24

48

Curing Time (hr)

Figure 6. Effect of emulsion type on pullout tension.

4

UW DATA worksheet

aggregate glass

emulsion

cure time

cure temp

response

st dev

sample size

st error

cov

CRS-2

2 hr

35

66.7

5.7

4

2.8

8.5%

CRS-2

6 hr

35

150.3

2.5

2

1.8

1.6%

CRS-2

24 hr

35

261.7

17.7

2

12.6

6.8%

CRS-2

2 hr

35

66.9

8.4

3

4.9

12.6%

CRS-2

6 hr

35

129.0

16.4

2

11.6

12.7%

35-30

CRS-2

24 hr

35

239.9

36.9

3

21.3

15.4%

CRS-2

CRS-2

2 hr

35

70.9

4.5

4

2.2

6.3%

granite

limestone

dolomite

glass

granite 45-30 CRS-2 limestone

dolomite

glass

granite 35-30 CRS-2P limestone

dolomite

glass

granite 45-30 CRS-2P limestone

dolomite

CRS-2

6 hr

35

127.6

6.3

4

3.2

5.0%

CRS-2

24 hr

35

252.0

9.7

4

4.9

3.9%

CRS-2

2 hr

35

81.1

4.2

3

2.4

5.2%

CRS-2

6 hr

35

116.7

10.0

3

5.8

8.6%

CRS-2

24 hr

35

195.8

11.2

4

5.6

5.7%

CRS-2

2 hr

45

93.4

8.9

3

5.1

9.5%

CRS-2

6 hr

45

118.0

12.8

4

6.4

10.8%

CRS-2

24 hr

45

210.2

21.2

3

12.3

10.1%

CRS-2

2 hr

45

73.8

6.7

2

4.7

9.1%

CRS-2

6 hr

45

102.1

5.9

3

3.4

5.8%

CRS-2

24 hr

45

222.4

13.8

4

6.9

6.2%

CRS-2

2 hr

45

58.9

7.1

4

3.5

12.0%

CRS-2

6 hr

45

99.8

2.4

3

1.4

2.4%

CRS-2

24 hr

45

220.7

14.3

4

7.2

6.5%

CRS-2

2 hr

45

95.3

4.7

4

2.3

4.9%

CRS-2

6 hr

45

171.2

6.3

4

3.1

3.7%

CRS-2

24 hr

45

245.8

20.6

3

11.9

8.4%

CRS-2P

2 hr

35

65.5

6.0

4

3.0

9.2%

CRS-2P

6 hr

35

73.6

6.6

3

3.8

9.0%

CRS-2P

24 hr

35

211.8

16.0

4

8.0

7.5%

CRS-2P

2 hr

35

65.7

11.7

3

6.8

17.8%

CRS-2P

6 hr

35

90.5

14.9

4

7.4

16.5%

CRS-2P

24 hr

35

184.8

25.4

4

12.7

13.8%

CRS-2P

2 hr

35

70.5

6.4

3

3.7

9.0%

CRS-2P

6 hr

35

171.7

51.1

2

36.2

29.8%

CRS-2P

24 hr

35

185.7

33.5

3

19.3

18.0%

CRS-2P

2 hr

35

57.7

3.3

4

1.6

5.7%

CRS-2P

6 hr

35

83.6

5.7

3

3.3

6.8%

CRS-2P

24 hr

35

166.1

14.8

3

8.5

8.9%

CRS-2P

2 hr

45

80.7

26.4

2

18.7

32.7%

CRS-2P

6 hr

45

103.8

12.1

4

6.0

11.6%

CRS-2P

24 hr

45

215.8

28.6

4

14.3

13.3%

CRS-2P

2 hr

45

61.3

7.2

4

3.6

11.7%

CRS-2P

6 hr

45

80.6

14.3

4

7.2

17.8%

CRS-2P

24 hr

45

152.7

30.6

3

17.7

20.1%

CRS-2P

2 hr

45

79.7

9.4

2

6.6

11.8%

CRS-2P

6 hr

45

112.3

14.3

4

7.1

12.7%

CRS-2P

24 hr

45

187.0

7.6

3

4.4

4.1%

CRS-2P

2 hr

45

89.5

9.4

4

4.7

10.5%

CRS-2P

6 hr

45

134.9

19.9

3

11.5

14.8%

CRS-2P

24 hr

45

247.4

8.6

3

4.9

3.5%

5

US DATA worksheet aggregate

emulsion

cure time cure temp response

st dev

sample size

st error

cov

CRS-65

2 hr

25

47.0

7.1

5

3.2

15.0%

CRS-65

6 hr

25

109.1

18.1

5

8.1

16.6%

CRS-65

24 hr

25

145.6

14.6

6

6.0

10.0%

CRS-65

2 hr

25

32.4

4.2

6

1.7

13.1%

CRS-65

6 hr

25

52.7

9.0

4

4.5

17.1%

CRS-65

24 hr

25

131.7

9.0

5

4.0

6.8%

CRS-65 + 3% latex

2 hr

25

39.4

5.8

3

3.3

14.7%

CRS-65 + 3% latex

6 hr

25

156.1

18.6

2

13.2

11.9%

CRS-65

CRS-65 + 3% latex

24 hr

25

120.9

6.8

3

3.9

5.6%

+ 3% Latex

CRS-65 + 3% latex

2 hr

25

35.6

3.9

3

2.2

10.9%

CRS-65 + 3% latex

6 hr

25

61.7

1.3

3

0.7

2.1%

CRS-65 + 3% latex

24 hr

25

147.5

5.4

3

3.1

3.6%

SS-66

2 hr

25

41.9

7.8

2

5.5

18.6%

SS-66

6 hr

25

56.3

4.2

3

2.4

7.5%

SS-66

24 hr

25

57.7

10.6

2

7.5

18.3%

SS-66

2 hr

25

43.0

8.5

3

4.9

19.7%

SS-66

6 hr

25

42.1

16.4

2

11.6

38.9%

SS-66

24 hr

25

98.7

11.4

2

8.1

11.5%

granite CRS-65 tillite

granite

tillite

granite SS-66 tillite

6

DATE:

02 November 2009

SUBJECT: RESULTS

UNIVERSITY OF STELLENBOSCH –FIRST TESTS- ANALYSIS &

TO:

EMULSION GROUP, UW-MADISON

FROM:

ANDRE GREYLING

INTRODUCTION A series of Bitumen Bond Strength (BBS) tests was conducted at the University of Stellenbosch using the new Patti Quantum Gold Equipment. The following emulsions were tested:   

Cationic Rapid Set Emulsion with 65% bitumen content. Anionic Spray Grade Emulsion ( Slow Set) with 60% bitumen content Cationic 65% with 3% Latex modified emulsion.

The following two aggregate types were used:  

Granite Tillite

The purpose of this set of experiments was to evaluate the new Patti Quantum apparatus, to investigate the repeatability of the tests and to check the results against theoretical expectations. The experimental factors examined in this series of tests are shown in Table 1. Variables 1 2 3

Aggregate Type Curing Temp Curing Interval

Details Granite 25°C 2,6 & 24 hr

Tillite 25°C 2,6 & 24 hr

Table 1: Experimental Factors and Levels

1

TEST SETUP & PREPARATION As this was the first test completed outside the UMW, special care was taken to copy the developed BBS procedure. There were however some problems experienced and these are discussed in detail below.

AGGREGATE PLATE PREPARATION The first action taken was to prepare suitable aggregate samples. Bulk hornfels aggregate samples, measuring 200x300mm on average, was collected from two commercial quarries in the Western Cape. These samples were delivered to a Geotechnical Laboratory to be saw cut into 20mm thick plates. This proved problematic due to the following reasons: (a)

In the greater Cape Town area there are no readily available mechanical feed rock saws. Most of the diamond blade saws available is the type that is used to saw concrete cores and has no clamps or screw driven feeding mechanism. Concrete cores usually have flat side that makes the manual feeding and cutting of these a fairly simple exercise.

(b)

The cutting of the samples was very time consuming and although possible, the finished samples had un parallel faces and uneven saw blade marks.

As these samples proved unpractical to work with an alternative arrangements had to be made. The South African National Road Agency keep Geological core samples of quarries used for surfacing seal aggregates and with the permission from SANRAL , four geological cores of two types of aggregate, granite and tillite, was collected from their storerooms. The cores are approximately 60mm in diameter and 1.0 to 1.5 meters in length. These cores were saw cut into 30mm thick discs on which the testing was completed.

LAPPING EQUIPMENT No lapping equipment is available at the University of Stellenbosch Engineering Department and the equipment in the Geological Department is typically used for 20mm diameter and smaller samples. There was however two unused lapping wheels, manufactured in the 1960‘s, available. These wheels was not able to lap the large rock samples but was suitable for the 60mm geological cores. The 80 grain grit supplied by the UMW was used to lap these samples to a suitable roughness.

2

ULTRASONIC CLEANER No ultrasonic cleaning equipment is available at the University of Stellenbosch Engineering Department. This was not critical as all samples were washed with clean water and left to dry. None of the samples have been reused in testing and the US is in the process of procuring an Ultrasonic Cleaner.

ENVIRONMENTAL CHAMBER No Environmental chambers were available and samples were therefore not cured in environmentally controlled environment.

PRACTICAL PROBLEMS EXPERIENCED The following additional practical problems were experienced:  Absence of an eye dropper caused some initial difficulty when pouring the emulsion. Syringes were used from the second test onwards.  The hornfels aggregate plate’s surfaces where not saw cut smooth enough and this caused the emulsion to soak underneath the silicone moulds.  The hornfels plates did not have parallel faces and required balancing during the curing phase to stop the emulsion from running off.  On the first test the problems was experienced with stubs that did not thread smoothly through the pressure ring. This lead to torsion failure of the samples. This was rectified by wire brushing the stubs and ensuring that they fit smoothly.  The loading rate dial is impractically sensitive and problems were experienced to get the rate adjusted accurately. After much trial and error it was eventually set at constant application rate of between 500 – 1000 kPa/s.

DEVIATION FROM PROCEDURES The following deviations form the BBS procedure was necessitated due to equipment shortfalls: (a) (b)

Smaller aggregate samples. No environmental chamber.

3

TEST RESULTS Results from the experiment are shown in the sections below.

TEST 1– Experimental Setup The experimental setup delivered no usable results.

TEST 2– CRS65 Set 1 The table below shows the results for the second test set. Please not that unfortunately no rate of application was recorded. Curing Time & Temp

Rock Type

Granite 2 hours @ 25°C Tillite

Granite 6 hours@ 25°C Tillite

Granite 24 hours @ 25°C

Tillite

Sample No.

Result [kPa]

Type of Failure

1

294.5

Adhesion

2

264.7

Adhesion/Cohesion

3

145.5

Adhesion/Cohesion

1

207.8

Adhesion

2

248.5

Adhesion

3

183.4

Adhesion

1

636.0

Cohesion

2

804.0

Cohesion

3

424.6

Cohesion

1

245.8

Adhesion

2

305.4

Adhesion

3

316.2

Adhesion

1

1023.6

Adhesion/Cohesion

2

1047.9

Adhesion/Cohesion

3

1113.0

Adhesion/Cohesion

1

934.1

Adhesion/Cohesion

2

977.5

Adhesion/Cohesion

3

901.6

Adhesion/Cohesion

4

TEST 3– CRS65 Set 2 The table below shows the results for the third test set. Curing Time

Rock Type




Sample No.

Result [kPa]

Rate [kPa/s]

Type of Failure

1

356.9

859.4

Adhesion

2

316.2

1033.8

Adhesion

3

386.7

693.8

Adhesion on steel

1

262.0

910.6

Adhesion/Cohesion

2

207.8

693.8

Adhesion

3

229.5

1338.0

Adhesion/Cohesion

1

644.1

1077.2

Adhesion on steel

2

936.8

1062.3

Adhesion on steel

3

741.7

1412.3

Cohesion

1

430.0

1033.8

Adhesion

2

402.9

743.3

Adhesion

3

248.5

1228.9

Adhesion

1

823.0

1077.2

Cohesion

2

963.9

983.3

Adhesion/Cohesion

3

1050.7

842.4

Adhesion/Cohesion

1

549.3

975.6

Cohesion

2

809.5

1033.8

Cohesion

3

917.9

715.5

Cohesion/Adhesion on steel

5

TEST 4– SS60 The table below shows the results for the fourth test set. Curing Time

Rock Type




Sample No.

Result [kPa]

Rate [kPa/s]

Type of Failure

1

-

-

FAIL

2

327.1

1181.5

Adhesion

3

251.2

867.2

Adhesion

1

229.5

693.8

Adhesion

2

337.9

910.6

Adhesion

3

321.6

693.8

Adhesion

1

421.9

650.4

Adhesion

2

373.1

983.3

Adhesion

3

370.4

1330.2

Adhesion

1

-

-

FAIL

2

210.5

838.7

Adhesion

3

370.4

1032.8

Adhesion

1

-

-

FAIL

2

449.0

1019.0

Adhesion

3

346.0

884.2

Adhesion

1

419.2

802.2

Adhesion

2

625.2

1012.1

Adhesion

3

736.3

1032.8

Adhesion

6

TEST 5– CRS65 + 3% Latex The table below shows the results for the fifth test set. Curing Time

Rock Type

Granite 2 hours Tillite



Sample No.

Result [kPa]

Rate [kPa/s]

Type of Failure

1

226.8

1077.2

Adhesion

2

286.4

1015.9

Adhesion/Cohesion

3

302.7

958.5

Adhesion on steel

1

275.6

792.9

Adhesion/Cohesion

2

237.6

823.9

Adhesion/Cohesion

3

224.1

892.0

Adhesion

1

514.1

1033.8

Adhesion on steel

2

1167.2

1120.5

Adhesion on steel

3

985.6

1142.2

Adhesion on steel

1

421.9

823.9

Adhesion

2

419.2

1164.6

Adhesion on steel

3

435.5

958.5

Adhesion

1

809.5

773.6

Cohesion

2

888.0

925.4

Cohesion

3

804.0

903.7

Cohesion

1

988.3

925.4

Cohesion

2

1004.6

933.7

Cohesion

3

1058.8

1107.2

Cohesion

7

TEST RESULTS DISCUSSION TEST 1- Experimental Setup Due to unforeseen circumstances no usable results was delivered by Test 1. The test did however serve to point out various practical points that were addressed in the further tests.

TEST 2– CRS65 Set 1 The bar charts below shows the results of Test 2 CRS65 SET 1 1200

TEST 1 TEST 2

1000

TEST 3

Tensile Strength (kPa)

Average 800

600

400

200

0 GRANITE 2

TILLITE 2

GRANITE 6

TILLITE 6

GRANITE 24

TILLITE 24

HOURS

HOURS

HOURS

HOURS

HOURS

HOURS

CURING TIME

The following observations can be made from this graph:  

Granite/CRS65 delivers the highest tensile strength after all three curing times. Tillite/CRS65 shows slower tensile strength development in the 2-6 hour range.

8

TEST 3– CRS65 Set 2 The bar charts below shows the results of Test Set 3 CRS65 SET 2 1200

TEST 1 TEST 2

1000

TEST 3


Average 800

600

400

200

0 GRANITE 2

TILLITE 2

GRANITE 6

TILLITE 6

GRANITE 24

TILLITE 24

HOURS

HOURS

HOURS

HOURS

HOURS

HOURS

CURING TIME

The following observations can be made from this graph: 

This test was a repeat of Test 2 and from the results it can be seen that there is an acceptable level of repeatability.

9

TEST 4– SS60 The table below shows the results for the fourth test set.

SS60 1200

TEST 1 TEST 2

1000

TEST 3


Average 800

600

400

200

0 GRANITE 2

TILLITE 2

GRANITE 6

TILLITE 6

GRANITE 24

TILLITE 24

HOURS

HOURS

HOURS

HOURS

HOURS

HOURS

CURING TIME




The test results show much lower tensile strengths development for both aggregates. This can be explained by the fact that Cationic emulsion is positively charged and acidic aggregates like Granite have high negative charges. These negative charges attract the positively charged cationic bitumen particles, leading to destabilisation of the surfactant system and subsequent coagulation of the bitumen particles. This breaking mechanism is absent when anionic emulsions are used with acidic aggregates, and the coagulation can only take place by evaporation of the water phase. Tillite will generally have a higher pH than Granite and this may explain the higher tensile strength development after 24 hours.

10

TEST 5– CRS65 + 3% Latex The bar chart below shows the results for the fifth test set.

CRS65/3% LATEX 1200

TEST 1 TEST 2

1000

TEST 3


Average 800

600

400

200

0 GRANITE 2

TILLITE 2

GRANITE 6

TILLITE 6

GRANITE 24

TILLITE 24

HOURS

HOURS

HOURS

HOURS

HOURS

HOURS

CURING TIME


The tensile strength after 6 hours is assumed to be a test anomaly and further tests will be done to investigate these results. Interesting to note is that Tillite after 24 hours deliver the highest tensile strength results.

11

EMULSION vs CURING TIME DISCUSSION RESULTS AFTER 2 HOURS @ 25°C CURING The bar chart below shows the combined emulsion and aggregate test results after two hours.

2 HOURS 1200

TEST 1 TEST 2

1000

TEST 3


Average 800

600

400

200

0 CRS S ET 1 CRS S ET 1 CRS S ET 2 CRS S ET 2 GRANITE

TILLITE

GRANITE

TILLITE

CRS /3

CRS /3

S S 60

S S 60

GRANITE

TILLITE

GRANITE

TILLITE

EMULSION TYPE

The following observations can be made from this graph: 

All samples shows limited development of tensile strength with CRS65/Granite and SS60/Tillite showing higher that average results.

12

RESULTS AFTER 6 HOURS @ 25°C CURING The bar chart below shows the combined emulsion and aggregate test results after six hours.

6 HOURS 1200

TES T 1 TES T 2

1000

TES T 3


Average 800

600

400

200


TILLITE

GRANITE

TILLITE

CRS /3

CRS /3

S S 60

S S 60

GRANITE

TILLITE

GRANITE

TILLITE

EMULSION TYPES


Granite samples show higher results than Tillite in combination with all emulsions. The results of CRS65/3 with granite may be a test anomaly.

13

RESULTS AFTER 24 HOURS @ 25°C CURING The bar chart below shows the combined emulsion and aggregate test results after twenty four hours. 24 HOURS 1200

TEST 1 TEST 2 TEST 3

1000


Average 800

600

400

200


TILLITE

GRANITE

TILLITE

CRS /3

CRS /3

S S 60

S S 60

GRANITE

TILLITE

GRANITE

TILLITE

EMULSION TYPES

The following observations can be made from this graph:   

Granite samples show higher results than Tillite in combination with the CRS65 emulsion. Tillite in combination with CRS65-3 shows higher results than Granite but still lower that the CRS65/Granite combinations. The SS60 Granite/Tillite combinations show a lower tensile strength that the Cationic Emulsions.

CONCLUSION This test phased served as an experimental phase to identify any practical problems than may be experienced. The prelim results show good repeatability and confirm some of the expected theoretical results. Further intensive testing will take place once a testing strategy is developed in conjunction with the UWM.

14


ADDENDUM 4- BBS TEST RESULTS OF TESTS COMPLETED IN 2010

122 AH GREYLING

December 2012

Rep 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Rate (kPa/s) 308.7410901 319.3341375 314.8890469 303.3243243 336.1734007 384.2788051 933.3119835 994.9028926 911.1556503 869.9216495 860.5206612 872.5683761 971.0106838 826.241453 994.1730769 697.777403

2 HOURS Tension (kPa) 294.539 264.728 256.598 356.87 316.219 386.681 467.983 484.243 432.752 427.332 427.332 419.202 454.433 386.681 465.273 413.782

Failure Type Adhesion Adhesion/Cohesion Adhesion/Cohesion Adhesion Adhesion Adhesion Steel Adhesion Adhesion/Cohesion Adhesion Adhesion Adhesion Adhesion Cohesion/Adhesion Cohesion Cohesion/Adhesion Cohesion/Adhesion

Rep 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

CRS65 on GRANITE 6 HOURS Rate (kPa/s) Tension (kPa) 299.3482143 603.486 293.9784278 804.031 327.387818 424.622 361.4685746 644.137 315.535197 936.824 258.4348497 335.19 842.6589717 803.054 898.1919097 954.778 904.7041252 636.007 1092.615385 511.344 1134.886364 549.285 1148.070362 538.445 836.8574879 692.918 797.5809179 660.397 716.0573356 589.936

Failure Type Cohesion Cohesion Cohesion Adhesion Steel Adhesion Steel Cohesion Cohesion Adhesion on Steel Cohesion Cohesion Cohesion Cohesion Adhesion on Aggregate Adhesion on Aggregate Adhesion/Cohesion

Rep 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Rate (kPa/s) 309.0416667 304.898458 311.0614869 394.7921627 369.4614795 384.1488117 442.581487 528.4945055 565.2824822 909.089192 985.8698846 897.4425612 952.4172462 973.7629758 899.044686

24 HOURS Tension (kPa) 1023.546 1047.936 1112.978 795.901 963.925 1050.647 898.883 817.581 874.492 866.362 939.534 953.084 684.788 562.835 744.409

Failure Type Adhesion/Cohesion Adhesion/Cohesion Adhesion/Cohesion Cohesion Adhesion Cohesion Adhesion Cohesion Adhesion Substrate Adhesion/Cohesion Adhesion/Cohesion Adhesion on Aggregate Adhesion Adhesion on Aggregate

Rep 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Rate (kPa/s) 273.897343 304.7487352 441.851602 412.5764463 489.3326226 959.4294872 819.8947368 858.18637 855.1944444 810.1260331 882.2643923 826.241453 716.0573356 641.8535354 682.9158249 770.9318182

2 HOURS Tension (kPa) 248.468 180.716 262.018 207.817 229.497 449.013 576.386 617.036 400.231 394.811 424.622 386.681 424.622 400.231 481.533 392.101

Failure Type Adhesion Adhesion Adhesion/Cohesion Adhesion Adhesion/Cohesion Adhesion Adhesion Adhesion Adhesion Adhesion Adhesion Adhesion Cohesion/Adhesion Cohesion Cohesion/Adhesion Cohesion/Adhesion

Rep 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

CRS65 on TILLITE 6 HOURS Rate (kPa/s) Tension (kPa) 207.0412805 245.758 283.4711438 305.379 331.8142707 316.219 371.9582754 272.858 391.7270531 402.942 345.5744089 248.468 961.2330918 795.901 945.0660377 901.593 849.3714586 809.451 925.2294686 766.09 977.9487909 687.498 1016.874396 841.972 912.1376812 755.25 826.6211962 787.77 926.2816189 549.285

Failure Type Adhesion Adhesion Adhesion Adhesion Adhesion Adhesion Cohesion Adhesion on Steel Cohesion Cohesion Cohesion Cohesion Cohesion Cohesion/Adhesion Steel Cohesion

Rep 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Rate (kPa/s) 291.6372151 284.3977306 305.8733823 381.9784423 378.0714619 386.4644211 921.7203842 820.8870056 946.0577125 860.7460651 951.4130435 943.2140609 907.2593383 923.6268657 950.6435432

24 HOURS Tension (kPa) 934.114 977.475 874.492 549.285 809.451 917.853 863.652 871.782 901.593 820.291 787.77 898.883 850.102 866.362 890.753

Failure Type Adhesion/Cohesion Adhesion/Cohesion Adhesion/Cohesion

Cohesion Cohesion Cohesion Cohesion Cohesion Cohesion Cohesion Cohesion Cohesion

2 HOURS Rep 1 2 3 4 5 6 7

Rate (kPa/s) 839.450253 823.7249135 798.3187184 775.4688027 766.3288364 615.5789474 583.0347705

Tension (kPa) Failure Type Rep 497.794 Adhesion on Aggregate 1 476.113 Adhesion on Aggregate 2 473.403 Adhesion on Aggregate 3 459.853 Adhesion on Aggregate 4 478.823 Adhesion on Aggregate 5 432.752 Adhesion on Aggregate 6 419.202Adhesion on Aggregate and Steel

SS60 ON GRANITE 6 HOURS Rate (kPa/s) 784.608769 789.1787521 774.1632997 650.8803894 661.8392603 636.689415

Tension (kPa) 470.693 467.983 459.853 467.983 465.273 457.143

24 HOURS Failure Type Rep Adhesion on Aggregate 1 Adhesion on Aggregate 2 Adhesion on Aggregate 3 Adhesion on Aggregate 4 Adhesion on Aggregate 5 Adhesion on Aggregate 6 7

Rate (kPa/s) 696.5348506 658.418637 715.7560386 729.767285 789.1787521 705.7272727 728.5387205

Tension (kPa) 495.084 473.403 668.528 462.563 467.983 427.332 478.823

Failure Type Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate

SS60 ON TILLITE 6 HOURS

2 HOURS Rep 1 2 3 4 5 6 7

Rate (kPa/s) 947.8461538 905.3140496 913.1025641 971.0106838 884.1495726 701.1649832 678.3535354

Tension (kPa) 443.592 438.172 430.042 454.433 413.782 416.492 402.942

Failure Type Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate Adhesion

Rep 1 2 3 4 5 6

Rate (kPa/s) 770.8988196 760.476431 755.9141414 766.3288364 955.0086505 988.5959596

Tension (kPa) 457.143 451.723 470.693 470.693 551.995 587.226

24 HOURS Failure Type Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate Adhesion on Aggregate

Rep 1 2 3 4 5 6 7

Rate (kPa/s) 702.6642512 692.3421419 846.4128015 883.6599327 869.4951321 933.8468013 810.1260331

Tension (kPa) Failure Type 755.25 Adhesion on Aggregate 598.066 Adhesion on Aggregate 912.433 Adhesion on Aggregate 524.894 Adhesion on Aggregate 625.167 Adhesion on Aggregate & Steel 554.705 Adhesion on Aggregate 576.386 Adhesion on Aggregate

CRS65-3% Latex on GRANITE 6 HOURS

2 HOURS Rep 1 2 3 4 5 6 7 8

Rate (kPa/s) 631.718663 398.3435327 430.5391181 616.6211699 548.8321256 617.816156 738.9072513 733.1010101

Tension (kPa) Failure Type 226.787 286.409 302.669 221.367 454.433 Adhesion on Steel 443.592 Adhesion on Aggregate 438.172 Adhesion on Steel 435.462 Adhesion on Steel

Rep 1 2 3 4 5 6

Rate (kPa/s) 216.4437895 414.7795423 377.7711767 1010.328502 977.5978261 869.2770199

Tension (kPa) Failure Type Rep 514.054 1 888.043 2 985.605 3 836.552 Adhesion 4 5 809.451 Adhesion on Steel 828.421 Adhesion on Aggregate 6

24 HOURS Rate (kPa/s) 378.2481308 254.1617478 375.7154206 974.3248792 944.458256 962.7540017

Tension (kPa) Failure Type 809.451 888.043 804.031 806.741 Adhesion on Aggregate 1018.126 Adhesion on Aggregate 1142.789 Adhesion on Aggregate

CRS65-3% Latex on TILLITE 6 HOURS

2 HOURS Rep 1 2 3 4 5 6 7 8

Rate (kPa/s) 463.9208754 506.6695096 477.7761194 463.9208754 720.6273187 957.3837953 899.5991471 976.8012821

Tension (kPa) 275.569 237.628 224.077 275.569 427.332 449.013 432.752 457.143

Failure Type

Rate (kPa/s) Tension (kPa) Failure Type 359.993 421.912 439.876 419.202 284.43 435.462 967.779 801.321 Cohesion Adhesion on Aggregate 993.963 823.001 Adhesion on Steel Adhesion on Steel 960.276 915.143 Adhesion on Steel Adhesion on Steel Adhesion on Steel

24 HOURS Rep 1 2 3 4 5 6

Rate (kPa/s) Tension (kPa) 395.326 988.315 404.256 1004.575 405.817 1058.777 742.961 974.765 881.356 1156.339 881.356 1156.339

Failure Type

Cohesion Cohesion Cohesion

SS60+ 3% Latex ON GRANITE 6 HOURS

2 HOURS Rep 1 2 3 4 5 6 7

Rate (kPa/s) 657.0497159 635.8038943 760.476431 746.7878788 665.9568846 673.4039829 657.9843528

Tension (kPa) 462.563 457.143 451.723 449.013 478.823 473.403 465.273

Failure Type Adhesion Adhesion Adhesion Adhesion Adhesion Adhesion Adhesion

Rep 1 2 3

Rate (kPa/s) 688.5730181 684.8038943 673.4951321 627.384058 607.7463768 594.6545894

Tension (kPa) 497.794 492.374 486.953 519.474 503.214 495.084

24 HOURS Failure Type Adhesion Adhesion Adhesion Adhesion Adhesion Adhesion

Rep 1 2

Rate (kPa/s) 933.8468013 944.8789986 1026.722814 1015.166311 1000.502066 965.2200855 1040.497863

Tension (kPa) 554.705 587.226 481.533 476.113 484.243 473.403 530.314


SS60 ON TILLITE 6 HOURS

2 HOURS Rep 1 2 3 4

Rate (kPa/s) 737.6632997 733.1010101 728.5387205 701.1649832 755.9141414 748.0472175 748.0472175

Tension (kPa) 438.172 438.172 432.752 419.202 457.143 449.013 443.592


Rep 1 2 3

Rate (kPa/s) 828.4134948 787.8501684 856.2861953 816.5986509 806.0993266 798.3187184

Tension (kPa) 481.533 467.983 508.634 486.953 484.243 481.533

24 HOURS Failure Type Adhesion Adhesion Adhesion Adhesion Adhesion Adhesion

Rep 1 2 3

Rate (kPa/s) 932.2878788 990.9285042 923.9786629 839.450253 941.1098748 966.3840683 988.5959596

Tension (kPa) 1107.558 1053.357 649.557 757.96 676.658 776.93 698.338


Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 1.062 0 1.187 0.125 1.312 0.25 1.422 0.36 1.547 0.485 1.672 0.61 1.781 0.719 1.906 0.844 2.016 0.954 2.141 1.079 2.25 1.188 2.375 1.313 2.5 1.438

294.539 308.7410901 Adhesion Tensile (kPa) 0 75.024 134.645 164.456 183.426 205.107 234.918 270.148 294.539 188.847 142.775 129.225 88.574

CRS65 - 2 Hours - Granite SET A - Completed 2009 25C Test2 Test3 Tensile Strength 264.728 Tensile Strength Application Rate 319.3341375 Application Rate Failure Type Adhesion/Cohesion Failure Type Time (Sec) Time (Sec) Tensile (kPa) Time (Sec) Time (Sec) 1.187 0 0 1.078 0 1.312 0.125 37.083 1.187 0.109 1.422 0.235 121.095 1.312 0.234 1.547 0.36 156.326 1.422 0.344 1.656 0.469 178.006 1.547 0.469 1.781 0.594 199.687 1.672 0.594 1.906 0.719 232.208 1.781 0.703 2.016 0.829 264.728 1.906 0.828 2.141 0.954 199.687 2.016 0.938 2.25 1.063 167.166 2.141 1.063 2.375 1.188 172.586 2.266 1.188 2.5 1.313 194.267 2.375 1.297 2.609 1.422 126.515 2.5 1.422

256.598 314.8890469 Adhesion/Cohesion Tensile (kPa) 0 83.154 137.355 164.456 186.136 202.397 221.367 237.628 251.178 256.598 245.758 229.497 215.947

CRS65 - 2 Hours - Granite SET B - Completed 2009 25C Test4 Test5 Test6 Tensile Strength 356.87 Tensile Strength 316.219 Tensile Strength 386.681 Application Rate 303.32432 Application Rate 336.1734 Application Rate 384.27881 Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Steel Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.953 0 0 1.078 0 0 0.719 0 0 1.062 0.109 93.994 1.203 0.125 129.225 0.828 0.109 72.314 1.187 0.234 169.876 1.312 0.234 183.426 0.953 0.234 159.036 1.297 0.344 205.107 1.437 0.359 215.947 1.062 0.343 196.977 1.422 0.469 234.918 1.562 0.484 205.107 1.187 0.468 229.497 1.547 0.594 234.918 1.672 0.594 199.687 1.312 0.593 253.888 1.656 0.703 213.237 1.797 0.719 199.687 1.422 0.703 270.148 1.781 0.828 221.367 1.906 0.828 210.527 1.547 0.828 234.918 1.891 0.938 240.338 2.031 0.953 229.497 1.672 0.953 245.758 2.016 1.063 270.148 2.156 1.078 256.598 1.781 1.062 272.858 2.141 1.188 297.249 2.266 1.188 286.409 1.906 1.187 299.959 2.25 1.297 324.35 2.391 1.313 316.219 2.016 1.297 329.77 2.375 1.422 356.87 2.5 1.422 251.178 2.141 1.422 362.291 2.484 1.531 340.61 2.625 1.547 245.758 2.25 1.531 386.681 2.609 1.656 188.847 2.734 1.656 243.048 2.375 1.656 329.77 2.719 1.766 142.775 2.859 1.781 234.918 2.5 1.781 321.64 2.844 1.891 104.835 2.984 1.906 234.918 2.609 1.89 316.219 2.969 2.016 75.024 3.094 2.016 229.497 2.734 2.015 310.799 3.078 2.125 47.923 3.219 2.141 224.077 2.844 2.125 305.379 3.203 2.25 31.663 3.328 2.25 218.657 2.969 2.25 299.959 3.312 2.359 18.113 3.453 2.375 215.947 3.094 2.375 294.539 3.437 2.484 7.272 3.578 2.5 210.527 3.203 2.484 291.829 3.562 2.609 1.852 3.687 2.609 207.817 3.328 2.609 205.107

CRS65 - 2 Hours - Granite SET 1 - Completed 2010 30C Test7 Test8 Test9 Tensile Strength 467.983 Tensile Strength 484.243 Tensile Strength 432.752 Application Rate 933.31198 Application Rate 994.90289 Application Rate 911.15565 Failure Type Adhesion Failure Type Adhesion/Cohesion Failure Type Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.953 0 0 0.703 0 0 0.703 0 0 1.078 0.125 316.219 0.828 0.125 110.255 0.812 0.109 275.569 1.187 0.234 416.492 0.937 0.234 351.45 0.937 0.234 389.391 1.312 0.359 467.983 1.062 0.359 424.622 1.062 0.359 432.752 1.437 0.484 451.723 1.187 0.484 481.533 1.172 0.469 427.332 1.547 0.594 449.013 1.297 0.594 484.243 1.297 0.594 427.332 1.672 0.719 449.013 1.422 0.719 462.563 1.406 0.703 430.042 1.781 0.828 446.303 1.547 0.844 465.273 1.531 0.828 430.042 1.906 0.953 449.013 1.656 0.953 473.403 1.656 0.953 432.752 2.031 1.078 449.013 1.781 1.078 484.243 1.766 1.063 430.042 2.141 1.188 264.728 1.891 1.188 495.084 1.891 1.188 0 2.016 1.313 324.35 2 1.297 0

CRS65 - 2 Hours - Granite SET 2 - Completed 2010 30C Test10 Test11 Test12 Tensile Strength 427.332 Tensile Strength 427.332 Tensile Strength 419.202 Application Rate 869.92165 Application Rate 860.52066 Application Rate 872.56838 Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.937 0 0 0.703 0 0 0.719 0 0 1.062 0.125 294.539 0.828 0.125 262.018 0.828 0.109 289.119 1.172 0.235 392.101 0.937 0.234 383.971 0.953 0.234 381.261 1.297 0.36 427.332 1.062 0.359 427.332 1.062 0.343 419.202 1.422 0.485 421.912 1.187 0.484 416.492 1.187 0.468 408.362 1.531 0.594 419.202 1.297 0.594 413.782 1.312 0.593 405.652 1.656 0.719 419.202 1.422 0.719 419.202 1.422 0.703 405.652 1.766 0.829 251.178 1.531 0.828 443.592 1.547 0.828 80.444 1.656 0.953 272.858 1.766 1.063 243.048 1.891 1.188 229.497 2.016 1.313 218.657

CRS65 - 2 Hours - Granite SET 3 - Completed 2010 30C -New Ring Test13 Test14 Test15 Test16 Tensile Strength 454.433 Tensile Strength 386.681 Tensile Strength 465.273 Tensile Strength 413.782 Application Rate 971.01068 Application Rate 826.24145 Application Rate 994.17308 Application Rate 697.7774 Failure Type Cohesion/Adhesion Failure Type Cohesion Failure Type Cohesion/Adhesion Failure Type Cohesion/Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.594 0 0 0.594 0 0 0.594 0 0 0.469 0 0 0.719 0.125 153.616 0.703 0.109 262.018 0.703 0.109 188.847 0.594 0.125 148.196 0.828 0.234 335.19 0.828 0.234 365.001 0.828 0.234 340.61 0.703 0.234 316.219 0.953 0.359 397.521 0.937 0.343 386.681 0.953 0.359 400.231 0.828 0.359 386.681 1.062 0.468 454.433 1.062 0.468 386.681 1.062 0.468 465.273 0.953 0.484 411.072 1.187 0.593 411.072 1.172 0.578 386.681 1.187 0.593 408.362 1.062 0.593 413.782 1.312 0.718 400.231 1.297 0.703 386.681 1.297 0.703 392.101 1.187 0.718 394.811 1.422 0.828 397.521 1.422 0.828 0 1.422 0.828 386.681 1.297 0.828 386.681 1.547 0.953 397.521 1.531 0.937 0 1.547 0.953 37.083 1.422 0.953 383.971 1.656 1.062 34.373 1.656 1.062 0 1.656 1.062 0 1.531 1.062 196.977

Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 1.078 0 1.187 0.109 1.312 0.234 1.437 0.359 1.547 0.469 1.672 0.594 1.781 0.703 1.906 0.828 2.016 0.938 2.141 1.063 2.266 1.188 2.375 1.297 2.5 1.422 2.625 1.547 2.734 1.656 2.859 1.781 2.969 1.891 3.094 2.016 3.219 2.141 3.328 2.25 3.453 2.375 3.562 2.484 3.687 2.609 3.797 2.719 3.922 2.844

603.486 299.3482143 Cohesion Tensile (kPa) 0 107.545 148.196 172.586 191.557 210.527 237.628 262.018 291.829 327.06 362.291 394.811 430.042 465.273 500.504 535.735 570.965 603.486 549.285 451.723 476.113 500.504 527.604 554.705 581.806

CRS65 - 6 Hours - Granite SET A - Completed 2009 25C Test2 Test3 Tensile Strength 804.031 Tensile Strength Application Rate 293.9784278 Application Rate Failure Type Cohesion Failure Type Time (Sec) Time (Sec) Tensile (kPa) Time (Sec) Time (Sec) 0.937 0 0 1.203 0 1.062 0.125 20.823 1.312 0.109 1.187 0.25 115.675 1.437 0.234 1.297 0.36 159.036 1.562 0.359 1.422 0.485 183.426 1.672 0.469 1.547 0.61 207.817 1.797 0.594 1.656 0.719 229.497 1.906 0.703 1.781 0.844 262.018 2.031 0.828 1.891 0.954 291.829 2.156 0.953 2.016 1.079 324.35 2.266 1.063 2.141 1.204 356.87 2.391 1.188 2.25 1.313 392.101 2.5 1.297 2.375 1.438 424.622 2.625 1.422 2.484 1.547 459.853 2.75 1.547 2.609 1.672 495.084 2.859 1.656 2.734 1.797 530.314 2.984 1.781 2.844 1.907 565.545 3.094 1.891 2.969 2.032 600.776 3.219 2.016 3.078 2.141 636.007 3.344 2.141 3.203 2.266 673.948 3.453 2.25 3.312 2.375 706.469 3.578 2.375 3.437 2.5 741.699 3.687 2.484 3.562 2.625 774.22 3.812 2.609 3.672 2.735 804.031 3.937 2.734 3.797 2.86 611.616 4.047 2.844

424.622 327.387818 Cohesion Tensile (kPa) 0 99.414 145.486 172.586 196.977 224.077 253.888 286.409 321.64 354.16 389.391 424.622 367.711 199.687 140.065 121.095 115.675 115.675 85.864 0 0 0 0 0 0

CRS65 - 6 Hours - Granite SET B - Completed 2009 25C Test4 Test5 Test6 Tensile Strength 644.137 Tensile Strength 936.824 Tensile Strength 335.19 Application Rate 361.46857 Application Rate 315.5352 Application Rate 258.43485 Failure Type Adhesion Steel Failure Type Adhesion Steel Failure Type Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 1.062 0 0 1.062 0 0 1.078 0 0 1.187 0.125 134.645 1.187 0.125 34.373 1.187 0.109 153.616 1.297 0.235 194.267 1.312 0.25 167.166 1.312 0.234 205.107 1.422 0.36 221.367 1.422 0.36 224.077 1.437 0.359 221.367 1.531 0.469 240.338 1.547 0.485 243.048 1.547 0.469 229.497 1.656 0.594 267.438 1.656 0.594 262.018 1.672 0.594 245.758 1.781 0.719 289.119 1.781 0.719 278.279 1.781 0.703 256.598 1.891 0.829 316.219 1.906 0.844 294.539 1.906 0.828 270.148 2.016 0.954 351.45 2.016 0.954 305.379 2.016 0.938 280.989 2.125 1.063 386.681 2.141 1.079 321.64 2.141 1.063 297.249 2.25 1.188 424.622 2.266 1.204 335.19 2.266 1.188 318.93 2.359 1.297 470.693 2.375 1.313 356.87 2.375 1.297 335.19 2.484 1.422 511.344 2.5 1.438 378.551 2.5 1.422 321.64 2.609 1.547 554.705 2.609 1.547 408.362 2.609 1.531 251.178 2.719 1.657 598.066 2.734 1.672 443.592 2.734 1.656 210.527 2.844 1.782 644.137 2.844 1.782 486.953 2.859 1.781 194.267 2.953 1.891 538.445 2.969 1.907 530.314 2.969 1.891 188.847 3.078 2.016 484.243 3.094 2.032 573.675 3.094 2.016 186.136 3.203 2.141 514.054 3.203 2.141 619.747 3.219 2.141 183.426 3.312 2.25 549.285 3.328 2.266 665.818 3.328 2.25 183.426 3.437 2.375 584.516 3.437 2.375 709.179 3.453 2.375 183.426 3.562 2.5 619.747 3.562 2.5 752.54 3.562 2.484 186.136 3.672 2.61 587.226 3.687 2.625 801.321 3.687 2.609 186.136 3.797 2.735 576.386 3.797 2.735 847.392 3.797 2.719 188.847 3.906 2.844 568.255 3.922 2.86 893.463 3.922 2.844 188.847 4.031 2.969 557.415 4.031 2.969 936.824 4.047 2.969 188.847 4.141 3.079 549.285 4.156 3.094 682.078 4.156 3.078 188.847 4.266 3.204 543.865 4.281 3.219 703.758 4.281 3.203 188.847 4.391 3.329 535.735 4.391 3.329 741.699 4.391 3.313 20.823

CRS65 - 6 Hours - Granite SET 1 - Completed 2010 30C Test7 Test8 Test9 Tensile Strength 803.054 Tensile Strength 954.778 Tensile Strength 636.007 Application Rate 842.65897 Application Rate 898.19191 Application Rate 904.70413 Failure Type Cohesion Failure Type Adhesion on Steel Failure Type Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.594 0 0 0.828 0 0 0.719 0 0 0.719 0.125 20.298 0.953 0.125 320.297 0.828 0.109 134.645 0.828 0.234 406.504 1.062 0.234 492.71 0.953 0.234 351.45 0.953 0.359 523.745 1.187 0.359 568.572 1.078 0.359 424.622 1.062 0.468 565.124 1.297 0.469 647.882 1.187 0.468 478.823 1.187 0.593 603.055 1.422 0.594 751.33 1.312 0.593 549.285 1.312 0.718 644.434 1.547 0.719 868.571 1.422 0.703 636.007 1.422 0.828 703.055 1.656 0.828 996.157 1.547 0.828 549.285 1.547 0.953 803.054 1.781 0.953 1120.295 1.672 0.953 592.646 1.656 1.062 758.227 1.891 1.063 954.778 1.781 1.062 654.977 1.781 1.187 758.227 2.016 1.188 1020.295 1.906 1.187 730.859 1.891 1.297 837.537 2.141 1.313 1116.846 2.016 1.297 798.611 2.016 1.422 934.088 2.25 1.422 1216.846 2.141 1.422 877.203 2.141 1.547 1027.191 2.375 1.547 1316.846 2.25 1.531 795.901 2.25 1.656 1123.743 2.484 1.656 1323.742 2.375 1.656 771.51 2.375 1.781 1196.156 2.609 1.781 1275.466 2.5 1.781 741.699 2.484 1.89 1085.812 2.734 1.906 1230.639 2.609 1.89 714.599

CRS65 - 6 Hours - Granite SET 2 - Completed 2010 30C Test10 Test11 Test12 Tensile Strength 511.344 Tensile Strength 549.285 Tensile Strength 538.445 Application Rate 1092.6154 Application Rate 1134.8864 Application Rate 1148.0704 Failure Type Cohesion Failure Type Cohesion Failure Type Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.594 0 0 0.828 0 0 0.828 0 0 0.703 0.109 58.764 0.953 0.125 291.829 0.953 0.125 340.61 0.828 0.234 381.261 1.062 0.234 438.172 1.062 0.234 451.723 0.937 0.343 462.563 1.187 0.359 495.084 1.187 0.359 495.084 1.062 0.468 511.344 1.312 0.484 549.285 1.297 0.469 538.445 1.187 0.593 500.504 1.422 0.594 508.634 1.422 0.594 511.344 1.297 0.703 500.504 1.547 0.719 503.214 1.531 0.703 495.084 1.422 0.828 503.214 1.656 0.828 500.504 1.656 0.828 519.474 1.547 0.953 503.214 1.781 0.953 500.504 1.781 0.953 570.965 1.656 1.062 503.214 1.906 1.078 64.184 1.891 1.063 636.007 1.781 1.187 253.888 2.016 1.188 0 2.016 1.188 486.953 1.891 1.297 0 2.141 1.313 0 2.125 1.297 465.273 2.016 1.422 0 2.25 1.422 0 2.25 1.422 443.592

CRS65 - 6 Hours - Granite SET 3 - Completed 2010 30C-New Ring Test13 Test14 Test15 Tensile Strength 692.918 Tensile Strength 660.397 Tensile Strength 589.936 Application Rate 836.85749 Application Rate 797.58092 Application Rate 716.05734 Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Adhesion/Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.484 0 0 0.344 0 0 0.469 0 0 0.594 0.11 91.284 0.469 0.125 56.053 0.594 0.125 102.125 0.719 0.235 324.35 0.578 0.234 310.799 0.703 0.234 321.64 0.844 0.36 397.521 0.703 0.359 392.101 0.828 0.359 386.681 0.953 0.469 446.303 0.828 0.484 435.462 0.937 0.468 411.072 1.078 0.594 516.764 0.937 0.593 486.953 1.062 0.593 424.622 1.187 0.703 598.066 1.062 0.718 568.255 1.187 0.718 438.172 1.312 0.828 692.918 1.172 0.828 660.397 1.297 0.828 465.273 1.422 0.938 587.226 1.297 0.953 757.96 1.422 0.953 500.504 1.547 1.063 465.273 1.422 1.078 630.587 1.531 1.062 570.965 1.672 1.188 432.752 1.531 1.187 470.693 1.656 1.187 589.936 1.781 1.297 424.622 1.656 1.312 432.752 1.766 1.297 446.303 1.906 1.422 39.793 1.766 1.422 421.912 1.891 1.422 416.492

Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.75 0 0.859 0.109 0.984 0.234 1.094 0.344 1.219 0.469 1.344 0.594 1.453 0.703 1.578 0.828 1.687 0.937 1.812 1.062 1.937 1.187 2.047 1.297 2.172 1.422 2.281 1.531 2.406 1.656 2.531 1.781 2.641 1.891 2.766 2.016 2.891 2.141 3 2.25 3.125 2.375 3.234 2.484 3.359 2.609 3.469 2.719 3.594 2.844 3.719 2.969 3.828 3.078 3.953 3.203 4.062 3.312 4.187 3.437 4.312 3.562

CRS65 - 24 Hours - Granite SET A - Completed 2009 25C Test2 Test3 1023.546 Tensile Strength 1047.936 Tensile Strength 309.0416667 Application Rate 304.898458 Application Rate Adhesion/Cohesion Failure Type Adhesion/Cohesion Failure Type Tensile (kPa) Time (Sec) Time (Sec) Tensile (kPa) Time (Sec) Time (Sec) 0 4.047 0 0 0.703 0 77.734 4.156 0.109 215.947 0.828 0.125 126.515 4.281 0.234 221.367 0.937 0.234 159.036 4.406 0.359 226.787 1.062 0.359 186.136 4.516 0.469 234.918 1.187 0.484 215.947 4.641 0.594 245.758 1.297 0.594 248.468 4.75 0.703 262.018 1.422 0.719 278.279 4.875 0.828 283.699 1.547 0.844 310.799 4.984 0.937 310.799 1.656 0.953 346.03 5.109 1.062 340.61 1.781 1.078 381.261 5.234 1.187 373.131 1.891 1.188 419.202 5.344 1.297 405.652 2.016 1.313 451.723 5.469 1.422 443.592 2.141 1.438 492.374 5.594 1.547 478.823 2.25 1.547 524.894 5.703 1.656 516.764 2.375 1.672 562.835 5.828 1.781 551.995 2.484 1.781 598.066 5.937 1.89 592.646 2.609 1.906 633.297 6.062 2.015 627.877 2.734 2.031 673.948 6.187 2.14 665.818 2.844 2.141 706.469 6.297 2.25 703.758 2.969 2.266 741.699 6.422 2.375 741.699 3.078 2.375 776.93 6.531 2.484 774.22 3.203 2.5 814.871 6.656 2.609 812.161 3.328 2.625 850.102 6.781 2.734 850.102 3.437 2.734 888.043 6.891 2.844 888.043 3.562 2.859 920.564 7.016 2.969 923.274 3.672 2.969 958.504 7.141 3.094 958.504 3.797 3.094 993.735 7.25 3.203 993.735 3.922 3.219 1023.546 7.375 3.328 1028.966 4.031 3.328 947.664 7.484 3.437 1047.936 4.156 3.453 779.64 7.609 3.562 771.51 4.281 3.578

1112.978 311.0614869 Adhesion/Cohesion Tensile (kPa) 0 64.184 121.095 153.616 183.426 215.947 248.468 280.989 318.93 351.45 392.101 427.332 462.563 500.504 538.445 573.675 611.616 649.557 687.498 722.729 760.67 795.901 833.842 869.072 907.013 944.954 980.185 1015.416 1050.647 1085.877 1112.978

CRS65 - 24 Hours - Granite SET B - Completed 2009 25C Test4 Test5 Test6 Tensile Strength 795.901 Tensile Strength 963.925 Tensile Strength 1050.647 Application Rate 394.79216 Application Rate 369.46148 Application Rate 384.14881 Failure Type Failure Type Failure Type Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 1.062 0 0 0.844 0 0 1.062 0 0 1.187 0.125 134.645 0.953 0.109 107.545 1.187 0.125 72.314 1.297 0.235 191.557 1.078 0.234 175.296 1.297 0.235 164.456 1.422 0.36 221.367 1.187 0.343 202.397 1.422 0.36 199.687 1.531 0.469 251.178 1.312 0.468 229.497 1.531 0.469 226.787 1.656 0.594 289.119 1.437 0.593 262.018 1.656 0.594 262.018 1.766 0.704 327.06 1.547 0.703 294.539 1.781 0.719 302.669 1.891 0.829 370.421 1.672 0.828 335.19 1.891 0.829 337.9 2.016 0.954 411.072 1.781 0.937 375.841 2.016 0.954 383.971 2.125 1.063 451.723 1.906 1.062 416.492 2.125 1.063 424.622 2.25 1.188 495.084 2.031 1.187 459.853 2.25 1.188 467.983 2.359 1.297 538.445 2.141 1.297 500.504 2.359 1.297 508.634 2.484 1.422 584.516 2.266 1.422 543.865 2.484 1.422 554.705 2.609 1.547 627.877 2.375 1.531 589.936 2.609 1.547 598.066 2.719 1.657 673.948 2.5 1.656 633.297 2.719 1.657 644.137 2.844 1.782 717.309 2.609 1.765 679.368 2.844 1.782 690.208 2.953 1.891 760.67 2.734 1.89 722.729 2.969 1.907 736.279 3.078 2.016 795.901 2.859 2.015 766.09 3.078 2.016 779.64 3.203 2.141 633.297 2.969 2.125 812.161 3.203 2.141 825.711 3.312 2.25 654.977 3.094 2.25 855.522 3.312 2.25 874.492 3.437 2.375 684.788 3.203 2.359 901.593 3.437 2.375 923.274 3.547 2.485 720.019 3.328 2.484 942.244 3.562 2.5 966.635 3.672 2.61 755.25 3.453 2.609 963.925 3.672 2.61 1012.706 3.781 2.719 790.481 3.562 2.718 738.989 3.797 2.735 1050.647 3.906 2.844 823.001 3.687 2.843 774.22 3.906 2.844 785.06 4.031 2.969 809.451 3.797 2.953 806.741 4.031 2.969 814.871 4.141 3.079 793.191 3.922 3.078 841.972 4.156 3.094 844.682 4.266 3.204 782.35 4.047 3.203 874.492 4.266 3.204 882.623 4.391 3.329 774.22 4.156 3.312 912.433 4.391 3.329 915.143 4.5 3.438 763.38 4.281 3.437 942.244 4.516 3.454 950.374 4.625 3.563 752.54 4.406 3.562 920.564 4.625 3.563 982.895

CRS65 - 24 Hours - Granite SET 1 - Completed 2010 30C Test7 Test8 Test9 Tensile Strength 898.883 Tensile Strength 817.581 Tensile Strength 874.492 Application Rate 442.58149 Application Rate 528.49451 Application Rate 565.28248 Failure Type Cohesion Failure Type Adhesion Failure Type Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.719 0 0 0.719 0 0 0.594 0 0 0.844 0.844 283.699 0.828 0.828 240.338 0.719 0.719 186.136 0.953 0.953 411.072 0.953 0.953 405.652 0.828 0.828 370.421 1.078 1.078 454.433 1.078 1.078 465.273 0.953 0.953 438.172 1.203 1.203 473.403 1.187 1.187 533.025 1.062 1.062 505.924 1.312 1.312 486.953 1.312 1.312 614.326 1.187 1.187 581.806 1.437 1.437 505.924 1.437 1.437 711.889 1.312 1.312 679.368 1.547 1.547 535.735 1.547 1.547 817.581 1.422 1.422 768.8 1.672 1.672 598.066 1.672 1.672 720.019 1.547 1.547 874.492 1.797 1.797 687.498 1.781 1.781 776.93 1.656 1.656 736.279 1.906 1.906 790.481 1.906 1.906 852.812 1.781 1.781 804.031 2.031 2.031 898.883 2.016 2.016 936.824 1.906 1.906 888.043 2.141 2.141 820.291 2.141 2.141 1018.126 2.016 2.016 966.635 2.266 2.266 831.131 2.266 2.266 1102.138 2.141 2.141 961.214 2.391 2.391 915.143 2.375 2.375 1034.386 2.25 2.25 923.274 2.5 2.5 993.735 2.5 2.5 996.445 2.375 2.375 890.753 2.625 2.625 1072.327 2.609 2.609 961.214 2.5 2.5 860.942 2.734 2.734 1161.759 2.734 2.734 928.694 2.609 2.609 831.131 2.859 2.859 1234.931 2.734 2.734 801.321 2.969 2.969 1145.499 2.844 2.844 774.22 3.094 3.094 1104.848 3.219 3.219 1066.907 3.328 3.328 1034.386

CRS65 - 24 Hours - Granite SET 2 - Completed 2010 30C Test10 Test11 Test12 Tensile Strength 866.362 Tensile Strength 939.534 Tensile Strength 953.084 Application Rate 909.08919 Application Rate 985.86988 Application Rate 897.44256 Failure Type Adhesion Substrate Failure Type Adhesion/Cohesion Failure Type Adhesion/Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.594 0 0 0.703 0 0 0.719 0 0 0.703 0.109 88.574 0.828 0.125 289.119 0.828 0.109 56.053 0.828 0.234 362.291 0.937 0.234 411.072 0.953 0.234 340.61 0.953 0.359 440.882 1.062 0.359 473.403 1.078 0.359 427.332 1.062 0.468 492.374 1.172 0.469 543.865 1.187 0.468 478.823 1.187 0.593 570.965 1.297 0.594 630.587 1.312 0.593 551.995 1.297 0.703 657.687 1.406 0.703 730.859 1.437 0.718 638.717 1.422 0.828 757.96 1.531 0.828 831.131 1.547 0.828 741.699 1.547 0.953 866.362 1.656 0.953 939.534 1.672 0.953 841.972 1.656 1.062 858.232 1.766 1.063 793.191 1.781 1.062 953.084 1.781 1.187 809.451 1.891 1.188 869.072 1.906 1.187 852.812 1.891 1.297 882.623 2 1.297 950.374 2.031 1.312 877.203 2.016 1.422 966.635 2.125 1.422 934.114 2.141 1.422 958.504 2.125 1.531 1045.226 2.25 1.547 893.463 2.266 1.547 1039.806 2.25 1.656 982.895 2.359 1.656 860.942 2.375 1.656 1096.718 2.375 1.781 947.664 2.5 1.781 1015.416 2.484 1.89 912.433 2.609 1.89 977.475 2.609 2.015 879.913 2.734 2.015 942.244 2.734 2.14 850.102

CRS65 - 24 Hours - Granite SET 3 - Completed 2010 30C New Ring Test13 Test14 Test15 Tensile Strength 684.788 Tensile Strength 562.835 Tensile Strength 744.409 Application Rate 952.41725 Application Rate 973.76298 Application Rate 899.04469 Failure Type Adhesion on Aggregate Failure Type Adhesion Failure Type Adhesion on Aggregate Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.5 0 0 0.469 0 0 0.594 0 0 0.625 0.125 134.645 0.578 0.109 96.704 0.703 0.109 305.379 0.734 0.234 356.87 0.703 0.234 329.77 0.828 0.234 402.942 0.859 0.359 438.172 0.812 0.343 421.912 0.937 0.343 438.172 0.969 0.469 508.634 0.937 0.468 486.953 1.062 0.468 486.953 1.094 0.594 589.936 1.047 0.578 562.835 1.172 0.578 560.125 1.219 0.719 684.788 1.172 0.703 546.575 1.297 0.703 654.977 1.328 0.828 600.776 1.297 0.828 462.563 1.422 0.828 744.409 1.453 0.953 489.664 1.406 0.937 446.303 1.531 0.937 514.054 1.562 1.062 465.273 1.531 1.062 443.592 1.656 1.062 459.853 1.687 1.187 457.143 1.641 1.172 438.172 1.766 1.172 232.208 1.812 1.312 286.409 1.766 1.297 18.113 1.891 1.297 0

CRS65 - 2 Hours - Tillite SET A - Completed 2009 25C Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 1.078 0 1.187 0.109 1.312 0.234 1.437 0.359 1.547 0.469 1.672 0.594 1.781 0.703 1.906 0.828 2.016 0.938 2.141 1.063 2.266 1.188 2.375 1.297 2.5 1.422

248.468 273.897343 Adhesion Tensile (kPa) 0 72.314 134.645 164.456 183.426 194.267 210.527 226.787 243.048 248.468 164.456 137.355 93.994

Test2 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 1.219 0 1.328 0.109 1.453 0.234 1.562 0.343 1.687 0.468 1.812 0.593 1.922 0.703 2.047 0.828 2.172 0.953 2.281 1.062 2.406 1.187 2.516 1.297 2.641 1.422 2.766 1.547 2.875 1.656

180.716 304.7487352 Adhesion Tensile (kPa) 0 45.213 129.225 159.036 175.296 180.716 178.006 172.586 169.876 167.166 164.456 164.456 167.166 169.876 169.876

CRS65 - 2 Hours - TILLITE SET A - Completed 2009 25C Test3 Test4 Test5 Tensile Strength 262.018 Tensile Strength 207.817 Tensile Strength 229.497 Application Rate 441.8516 Application Rate 412.57645 Application Rate 489.33262 Failure Type Adhesion/Cohesion Failure Type Adhesion Failure Type Adhesion/Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.844 0 0 0.703 0 0 1.312 0 0 0.953 0.109 53.343 0.828 0.125 56.053 1.422 0.235 153.616 1.078 0.234 167.166 0.953 0.25 142.775 1.547 0.36 207.817 1.187 0.343 207.817 1.062 0.359 180.716 1.656 0.469 229.497 1.312 0.468 234.918 1.187 0.484 199.687 1.781 0.594 215.947 1.437 0.593 262.018 1.297 0.594 207.817 1.906 0.719 199.687 1.547 0.703 215.947 1.422 0.719 205.107 2.016 0.829 194.267 1.672 0.828 191.557 1.547 0.844 191.557 2.141 0.954 191.557 1.781 0.937 183.426 1.656 0.953 186.136 2.25 1.063 194.267 1.906 1.062 180.716 1.781 1.078 183.426 2.375 1.188 194.267 2.031 1.187 180.716 1.891 1.188 183.426 2.484 1.297 191.557 2.141 1.297 180.716 2.016 1.313 183.426 2.609 1.422 194.267 2.266 1.422 180.716 2.141 1.438 186.136 2.734 1.547 134.645 2.375 1.531 180.716 2.25 1.547 186.136 2.844 1.657 1.852 2.5 1.656 180.716 2.375 1.672 186.136

CRS65 - 2 Hours - TILLITE SET 1 - Completed 2010 30C Test6 Test7 Test8 Tensile Strength 449.013 Tensile Strength 576.386 Tensile Strength 617.036 Application Rate 959.42949 Application Rate 819.89474 Application Rate 858.18637 Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.594 0 0 0.703 0 0 0.703 0 0 0.703 0.109 102.125 0.828 0.125 112.965 0.828 0.125 129.225 0.828 0.234 335.19 0.937 0.234 346.03 0.953 0.25 335.19 0.937 0.343 408.362 1.062 0.359 413.782 1.062 0.359 402.942 1.062 0.468 449.013 1.172 0.469 470.693 1.187 0.484 467.983 1.187 0.593 430.042 1.297 0.594 549.285 1.297 0.594 541.155 1.297 0.703 424.622 1.406 0.703 576.386 1.422 0.719 617.036 1.422 0.828 424.622 1.531 0.828 538.445 1.531 0.828 535.735 1.531 0.937 424.622 1.656 0.953 595.356 1.656 0.953 592.646 1.656 1.062 424.622 1.766 1.063 660.397 1.781 1.078 657.687 1.781 1.187 37.083 1.891 1.188 736.279 1.891 1.188 725.439 1.891 1.297 0 2 1.297 698.338 2.016 1.313 779.64 2.016 1.422 0 2.125 1.422 665.818 2.125 1.422 690.208 2.25 1.547 641.427 2.25 1.547 663.108 2.359 1.656 617.036 2.359 1.656 638.717

CRS65 - 2 Hours - TILLITE SET 2 - Completed 2010 30C Test9 Test10 Test11 Test12 Tensile Strength 400.231 Tensile Strength 394.811 Tensile Strength 424.622 Tensile Strength 386.681 Application Rate 855.19444 Application Rate 810.12603 Application Rate 882.26439 Application Rate 826.24145 Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.594 0 0 0.594 0 0 0.484 0 0 0.719 0 0 0.703 0.109 104.835 0.719 0.125 129.225 0.594 0.11 278.279 0.828 0.109 286.409 0.828 0.234 316.219 0.844 0.25 327.06 0.719 0.235 367.711 0.953 0.234 373.131 0.937 0.343 397.521 0.953 0.359 389.391 0.844 0.36 424.622 1.078 0.359 381.261 1.062 0.468 400.231 1.078 0.484 392.101 0.953 0.469 413.782 1.187 0.468 386.681 1.187 0.593 400.231 1.187 0.593 394.811 1.078 0.594 451.723 1.312 0.593 386.681 1.297 0.703 400.231 1.312 0.718 394.811 1.187 0.703 505.924 1.422 0.703 85.864 1.422 0.828 397.521 1.422 0.828 343.32 1.312 0.828 438.172 1.531 0.937 140.065 1.422 0.938 416.492 1.656 1.062 123.805 1.547 1.063 397.521 1.781 1.187 115.675 1.672 1.188 381.261 1.891 1.297 104.835 1.781 1.297 365.001

CRS65 - 2 Hours - TILLITE SET 3 - Completed 2010 30C - New Ring Test13 Test14 Test15 Test16 Tensile Strength 424.622 Tensile Strength 400.231 Tensile Strength 481.533 Tensile Strength 392.101 Application Rate 716.05734 Application Rate 641.85354 Application Rate 682.91582 Application Rate 770.93182 Failure Type Cohesion/Adhesion Failure Type Cohesion Failure Type Cohesion/Adhesion Failure Type Cohesion/Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.469 0 0 0.484 0 0 0.484 0 0 0.469 0 0 0.594 0.125 47.923 0.594 0.11 102.125 0.594 0.11 251.178 0.594 0.125 248.468 0.703 0.234 286.409 0.719 0.235 305.379 0.719 0.235 351.45 0.719 0.25 346.03 0.828 0.359 367.711 0.828 0.344 373.131 0.828 0.344 411.072 0.828 0.359 392.101 0.937 0.468 419.202 0.953 0.469 400.231 0.953 0.469 481.533 0.953 0.484 373.131 1.062 0.593 424.622 1.078 0.594 381.261 1.078 0.594 405.652 1.062 0.593 362.291 1.172 0.703 386.681 1.187 0.703 375.841 1.187 0.703 375.841 1.187 0.718 362.291 1.297 0.828 375.841 1.312 0.828 373.131 1.312 0.828 370.421 1.297 0.828 262.018 1.422 0.953 370.421 1.422 0.938 370.421 1.422 0.938 148.196 1.422 0.953 0

Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 1.172 0 1.297 0.125 1.406 0.234 1.531 0.359 1.656 0.484 1.766 0.594 1.891 0.719 2 0.828 2.125 0.953 2.25 1.078 2.359 1.187 2.484 1.312 2.594 1.422 2.719 1.547 2.844 1.672 2.953 1.781

245.758 207.0412805 Adhesion Tensile (kPa) 0 93.994 131.935 153.616 169.876 180.716 191.557 202.397 215.947 237.628 245.758 175.296 134.645 123.805 118.385 118.385

CRS65 - 6 Hours - Tillite SET A - Completed 2009 25C Test2 Tensile Strength 305.379 Application Rate 283.4711438 Failure Type Adhesion Time (Sec) Time (Sec) Tensile (kPa) 1.203 0 0 1.328 0.125 39.793 1.437 0.234 123.805 1.562 0.359 161.746 1.672 0.469 186.136 1.797 0.594 205.107 1.922 0.719 229.497 2.031 0.828 251.178 2.156 0.953 270.148 2.266 1.063 289.119 2.391 1.188 305.379 2.5 1.297 240.338 2.625 1.422 156.326 2.75 1.547 129.225 2.859 1.656 123.805 2.984 1.781 121.095

Test3 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 1.172 0 1.297 0.125 1.406 0.234 1.531 0.359 1.656 0.484 1.766 0.594 1.891 0.719 2 0.828 2.125 0.953 2.25 1.078 2.359 1.187 2.484 1.312 2.594 1.422 2.719 1.547 2.844 1.672 2.953 1.781


CRS65 - 6 Hours - Tillite SET B - Completed 2009 25C Test4 Test5 Test6 Tensile Strength 272.858 Tensile Strength 402.942 Tensile Strength 248.468 Application Rate 371.95828 Application Rate 391.72705 Application Rate 345.57441 Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.937 0 0 0.953 0 0 1.187 0 0 1.062 0.125 129.225 1.078 0.125 72.314 1.312 0.125 153.616 1.187 0.25 199.687 1.187 0.234 153.616 1.422 0.235 196.977 1.297 0.36 226.787 1.312 0.359 194.267 1.547 0.36 218.657 1.422 0.485 245.758 1.422 0.469 221.367 1.672 0.485 229.497 1.531 0.594 272.858 1.547 0.594 253.888 1.781 0.594 240.338 1.656 0.719 267.438 1.672 0.719 286.409 1.906 0.719 248.468 1.781 0.844 226.787 1.781 0.828 324.35 2.016 0.829 210.527 1.891 0.954 234.918 1.906 0.953 362.291 2.141 0.954 199.687 2.016 1.079 259.308 2.031 1.078 402.942 2.266 1.079 196.977 2.125 1.188 283.699 2.141 1.188 302.669 2.375 1.188 199.687 2.25 1.313 313.509 2.266 1.313 226.787 2.5 1.313 199.687 2.359 1.422 343.32 2.375 1.422 199.687 2.609 1.422 196.977 2.484 1.547 373.131 2.5 1.547 188.847 2.734 1.547 196.977 2.609 1.672 408.362 2.609 1.656 183.426 2.844 1.657 194.267 2.719 1.782 430.042 2.734 1.781 180.716 2.969 1.782 1.852

CRS65 - 6 Hours - Tillite SET 1 - Completed 2010 30C Test7 Test8 Test9 Tensile Strength 795.901 Tensile Strength 901.593 Tensile Strength 809.451 Application Rate 961.23309 Application Rate 945.06604 Application Rate 849.37146 Failure Type Cohesion Failure Type Adhesion on Steel Failure Type Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 1.172 0 0 0.937 0 20.823 0.828 0 0 1.297 0.125 299.959 1.062 0.125 310.799 0.937 0.109 159.036 1.406 0.234 402.942 1.172 0.235 397.521 1.062 0.234 351.45 1.531 0.359 457.143 1.297 0.36 459.853 1.172 0.344 411.072 1.656 0.484 522.184 1.422 0.485 533.025 1.297 0.469 465.273 1.766 0.594 608.906 1.531 0.594 617.036 1.422 0.594 538.445 1.891 0.719 698.338 1.656 0.719 706.469 1.531 0.703 617.036 2 0.828 795.901 1.766 0.829 806.741 1.656 0.828 711.889 2.125 0.953 779.64 1.891 0.954 901.593 1.781 0.953 809.451 2.25 1.078 736.279 2.016 1.079 1012.706 1.891 1.063 684.788 2.359 1.187 801.321 2.125 1.188 1104.848 2.016 1.188 749.83 2.484 1.312 877.203 2.25 1.313 901.593 2.125 1.297 823.001 2.594 1.422 955.794 2.359 1.422 969.345 2.25 1.422 893.463 2.719 1.547 953.084 2.484 1.547 1045.226 2.359 1.531 812.161 2.844 1.672 917.853 2.609 1.672 1126.528 2.484 1.656 782.35 2.953 1.781 888.043 2.719 1.782 1134.659 2.609 1.781 755.25

CRS65 - 6 Hours - Tillite SET 2 - Completed 2010 30C Test10 Test11 Test12 Tensile Strength 766.09 Tensile Strength 687.498 Tensile Strength 841.972 Application Rate 925.22947 Application Rate 977.94879 Application Rate 1016.8744 Failure Type Cohesion Failure Type Cohesion Failure Type Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.719 0 0 0.828 0 0 0.844 0 0 0.828 0.109 215.947 0.937 0.109 118.385 0.953 0.109 286.409 0.953 0.234 411.072 1.062 0.234 373.131 1.078 0.234 416.492 1.062 0.343 470.693 1.187 0.359 459.853 1.203 0.359 476.113 1.187 0.468 503.214 1.297 0.469 516.764 1.312 0.468 541.155 1.312 0.593 570.965 1.422 0.594 595.356 1.437 0.593 636.007 1.422 0.703 660.397 1.531 0.703 687.498 1.547 0.703 730.859 1.547 0.828 766.09 1.656 0.828 641.427 1.672 0.828 841.972 1.656 0.937 673.948 1.766 0.938 682.078 1.781 0.937 720.019 1.781 1.062 736.279 1.891 1.063 757.96 1.906 1.062 793.191 1.906 1.187 817.581 2.016 1.188 660.397 2.031 1.187 869.072 2.016 1.297 744.409 2.125 1.297 636.007 2.141 1.297 950.374 2.141 1.422 711.889 2.25 1.422 608.906 2.266 1.422 855.522 2.266 1.547 684.788 2.359 1.531 587.226 2.375 1.531 825.711 2.375 1.656 660.397 2.484 1.656 568.255 2.5 1.656 793.191 2.609 1.781 543.865

CRS65 - 6 Hours - Tillite SET 3 - Completed 2010 30C -New Ring Test13 Test14 Test15 Tensile Strength 755.25 Tensile Strength 787.77 Tensile Strength 549.285 Application Rate 912.13768 Application Rate 826.6212 Application Rate 926.28162 Failure Type Cohesion Failure Type Cohesion/Adhesion Failure Steel Type Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.594 0 0 0.469 0 0 0.594 0 0 0.719 0.125 253.888 0.594 0.125 39.793 0.719 0.125 278.279 0.828 0.234 367.711 0.703 0.234 297.249 0.828 0.234 370.421 0.953 0.359 421.912 0.828 0.359 381.261 0.953 0.359 419.202 1.062 0.468 478.823 0.937 0.468 438.172 1.062 0.468 478.823 1.187 0.593 562.835 1.062 0.593 519.474 1.187 0.593 549.285 1.312 0.718 652.267 1.187 0.718 603.486 1.297 0.703 505.924 1.422 0.828 755.25 1.297 0.828 692.918 1.422 0.828 416.492 1.547 0.953 576.386 1.422 0.953 787.77 1.547 0.953 394.811 1.656 1.062 449.013 1.547 1.078 514.054 1.656 1.062 386.681 1.781 1.187 416.492 1.656 1.187 432.752 1.781 1.187 386.681 1.906 1.312 12.692 1.781 1.312 411.072 1.891 1.297 37.083

Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.734 0 0.844 0.11 0.969 0.235 1.078 0.344 1.203 0.469 1.328 0.594 1.437 0.703 1.562 0.828 1.687 0.953 1.797 1.063 1.922 1.188 2.031 1.297 2.156 1.422 2.266 1.532 2.391 1.657 2.516 1.782 2.625 1.891 2.75 2.016 2.859 2.125 2.984 2.25 3.109 2.375 3.219 2.485 3.344 2.61 3.453 2.719 3.578 2.844 3.703 2.969 3.812 3.078 3.937 3.203 4.062 3.328 4.172 3.438 4.297 3.563

CRS65 - 24 Hours - Tillite SET A - Completed 2009 25C Test2 Test3 934.114 Tensile Strength 977.475 Tensile Strength 291.6372151 Application Rate 284.3977306 Application Rate Adhesion/Cohesion Failure Type Adhesion/Cohesion Failure Type Tensile (kPa) Time (Sec) Time (Sec) Tensile (kPa) Time (Sec) Time (Sec) 0 0.719 0 0 0.953 0 50.633 0.828 0.109 47.923 1.062 0.109 104.835 0.953 0.234 104.835 1.187 0.234 140.065 1.078 0.359 140.065 1.312 0.359 164.456 1.187 0.468 167.166 1.422 0.469 194.267 1.312 0.593 188.847 1.547 0.594 224.077 1.422 0.703 210.527 1.656 0.703 256.598 1.547 0.828 232.208 1.781 0.828 289.119 1.672 0.953 259.308 1.906 0.953 324.35 1.781 1.062 291.829 2.016 1.063 356.87 1.906 1.187 324.35 2.141 1.188 389.391 2.031 1.312 356.87 2.266 1.313 424.622 2.141 1.422 392.101 2.375 1.422 457.143 2.266 1.547 427.332 2.5 1.547 492.374 2.375 1.656 459.853 2.609 1.656 527.604 2.5 1.781 497.794 2.734 1.781 562.835 2.609 1.89 527.604 2.859 1.906 595.356 2.734 2.015 568.255 2.969 2.016 630.587 2.859 2.14 600.776 3.094 2.141 665.818 2.969 2.25 636.007 3.219 2.266 701.048 3.094 2.375 671.238 3.328 2.375 733.569 3.219 2.5 706.469 3.453 2.5 771.51 3.328 2.609 741.699 3.562 2.609 804.031 3.453 2.734 776.93 3.687 2.734 839.262 3.562 2.843 812.161 3.812 2.859 874.492 3.687 2.968 847.392 3.922 2.969 907.013 3.812 3.093 879.913 4.047 3.094 934.114 3.922 3.203 915.143 4.156 3.203 795.901 4.047 3.328 947.664 4.281 3.328 711.889 4.156 3.437 977.475 4.406 3.453 736.279 4.281 3.562 850.102 4.516 3.563

874.492 305.8733823 Adhesion/Cohesion Tensile (kPa) 0 83.154 129.225 161.746 188.847 221.367 251.178 286.409 318.93 351.45 389.391 421.912 457.143 492.374 527.604 562.835 598.066 638.717 671.238 706.469 741.699 776.93 812.161 844.682 874.492 690.208 682.078 709.179 736.279 766.09 795.901

CRS65 - 24 Hours - Tillite SET B - Completed 2009 25C Test4 Test5 Test6 Tensile Strength 549.285 Tensile Strength 809.451 Tensile Strength 917.853 Application Rate 381.97844 Application Rate 378.07146 Application Rate 386.46442 Failure Type Failure Type Failure Type Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.703 0 0 0.953 0 0 0.844 0 0 0.828 0.125 37.083 1.078 0.125 129.225 0.969 0.125 77.734 0.953 0.25 159.036 1.187 0.234 186.136 1.094 0.25 167.166 1.062 0.359 202.397 1.312 0.359 218.657 1.203 0.359 207.817 1.187 0.484 229.497 1.437 0.484 248.468 1.328 0.484 240.338 1.312 0.609 262.018 1.547 0.594 275.569 1.453 0.609 272.858 1.422 0.719 299.959 1.672 0.719 305.379 1.562 0.718 308.089 1.547 0.844 335.19 1.781 0.828 335.19 1.687 0.843 351.45 1.656 0.953 381.261 1.906 0.953 375.841 1.797 0.953 392.101 1.781 1.078 419.202 2.016 1.063 419.202 1.922 1.078 435.462 1.891 1.188 462.563 2.141 1.188 462.563 2.047 1.203 478.823 2.016 1.313 505.924 2.266 1.313 505.924 2.156 1.312 524.894 2.141 1.438 549.285 2.375 1.422 549.285 2.281 1.437 568.255 2.25 1.547 451.723 2.5 1.547 589.936 2.391 1.547 608.906 2.375 1.672 283.699 2.609 1.656 638.717 2.516 1.672 654.977 2.484 1.781 221.367 2.734 1.781 684.788 2.641 1.797 698.338 2.609 1.906 196.977 2.859 1.906 725.439 2.75 1.906 744.409 2.734 2.031 188.847 2.969 2.016 771.51 2.875 2.031 787.77 2.844 2.141 186.136 3.094 2.141 809.451 2.984 2.14 836.552 2.969 2.266 180.716 3.203 2.25 638.717 3.109 2.265 879.913 3.078 2.375 183.426 3.328 2.375 598.066 3.219 2.375 917.853 3.203 3.437 2.484 633.297 3.344 2.5 709.179 3.312 3.562 2.609 665.818 3.469 2.625 722.729 3.437 3.687 2.734 701.048 3.578 2.734 755.25

CRS65 - 24 Hours - Tillite SET 1 - Completed 2010 30C Test7 Test8 Test9 Tensile Strength 863.652 Tensile Strength 871.782 Tensile Strength 901.593 Application Rate 921.72038 Application Rate 820.88701 Application Rate 946.05771 Failure Type Cohesion Failure Type Cohesion Failure Type Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.719 0 0 0.719 0 0 0.828 0 0 0.828 0.109 126.515 0.828 0.109 112.965 0.937 0.109 256.598 0.953 0.234 351.45 0.953 0.234 351.45 1.062 0.234 392.101 1.062 0.343 430.042 1.078 0.359 424.622 1.187 0.359 451.723 1.187 0.468 492.374 1.187 0.468 486.953 1.297 0.469 524.894 1.312 0.593 568.255 1.312 0.593 557.415 1.422 0.594 606.196 1.422 0.703 665.818 1.422 0.703 644.137 1.547 0.719 701.048 1.547 0.828 760.67 1.547 0.828 736.279 1.656 0.828 801.321 1.656 0.937 863.652 1.656 0.937 836.552 1.781 0.953 901.593 1.781 1.062 736.279 1.781 1.062 871.782 1.891 1.063 790.481 1.906 1.187 801.321 1.906 1.187 782.35 2.016 1.188 833.842 2.016 1.297 877.203 2.016 1.297 858.232 2.125 1.297 915.143 2.141 1.422 958.504 2.141 1.422 936.824 2.25 1.422 993.735 2.25 1.531 907.013 2.25 1.531 1009.996 2.375 1.547 1072.327 2.375 1.656 874.492 2.375 1.656 915.143 2.484 1.656 988.315 2.5 1.781 841.972 2.5 1.781 882.623 2.609 1.781 953.084 2.609 1.89 814.871 2.609 1.89 850.102 2.719 1.891 917.853 2.734 2.015 785.06 2.734 2.015 823.001 2.844 2.016 888.043

CRS65 - 24 Hours - Tillite SET 2 - Completed 2010 30C Test10 Test11 Test12 Tensile Strength 820.291 Tensile Strength 787.77 Tensile Strength 898.883 Application Rate 860.74607 Application Rate 951.41304 Application Rate 943.21406 Failure Type Cohesion Failure Type Cohesion Failure Type Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.578 0 0 0.703 0 0 0.703 0 0 0.703 0.125 186.136 0.828 0.125 229.497 0.828 0.125 297.249 0.828 0.25 370.421 0.937 0.234 386.681 0.937 0.234 397.521 0.937 0.359 440.882 1.062 0.359 446.303 1.062 0.359 459.853 1.062 0.484 508.634 1.172 0.469 511.344 1.187 0.484 533.025 1.172 0.594 587.226 1.297 0.594 589.936 1.297 0.594 622.457 1.297 0.719 684.788 1.406 0.703 690.208 1.422 0.719 720.019 1.422 0.844 776.93 1.531 0.828 787.77 1.531 0.828 825.711 1.531 0.953 820.291 1.656 0.953 722.729 1.656 0.953 898.883 1.656 1.078 747.12 1.766 1.063 755.25 1.766 1.063 771.51 1.766 1.188 823.001 1.891 1.188 831.131 1.891 1.188 841.972 1.891 1.313 904.303 2 1.297 909.723 2.016 1.313 923.274 2 1.422 982.895 2.125 1.422 923.274 2.125 1.422 999.155 2.125 1.547 936.824 2.25 1.547 882.623 2.25 1.547 985.605 2.25 1.672 898.883 2.359 1.656 850.102 2.359 1.656 947.664 2.484 1.781 823.001 2.484 1.781 915.143 2.609 1.906 790.481 2.609 1.906 882.623 2.719 2.016 763.38 2.719 2.016 852.812

CRS65 - 24 Hours - Tillite SET 3 - Completed 2010 30C -New Ring Test13 Test14 Test15 Tensile Strength 850.102 Tensile Strength 866.362 Tensile Strength 890.753 Application Rate 907.25934 Application Rate 923.62687 Application Rate 950.64354 Failure Type Cohesion Failure Type Cohesion Failure Type Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.594 0 0 0.484 0 0 0.469 0 0 0.703 0.109 102.125 0.594 0.11 156.326 0.578 0.109 240.338 0.828 0.234 335.19 0.719 0.235 351.45 0.703 0.234 375.841 0.937 0.343 413.782 0.828 0.344 424.622 0.812 0.343 435.462 1.062 0.468 476.113 0.953 0.469 489.664 0.937 0.468 503.214 1.172 0.578 554.705 1.078 0.594 576.386 1.062 0.593 589.936 1.297 0.703 646.847 1.187 0.703 668.528 1.172 0.703 684.788 1.422 0.828 749.83 1.312 0.828 766.09 1.297 0.828 787.77 1.531 0.937 850.102 1.422 0.938 866.362 1.406 0.937 890.753 1.656 1.062 654.977 1.547 1.063 543.865 1.531 1.062 617.036 1.766 1.172 486.953 1.672 1.188 459.853 1.656 1.187 478.823 1.891 1.297 449.013 1.781 1.297 432.752 1.766 1.297 443.592 2.016 1.422 1.852 1.906 1.422 0 1.891 1.422 289.119

CRS65 3% - 2 Hours - Granite SET D - Completed 2009 25C Test1 Tensile Strength 226.787 Application Rate 631.718663 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.953 0 0 1.078 0.125 134.645 1.187 0.234 191.557 1.312 0.359 226.787 1.422 0.469 215.947 1.547 0.594 207.817 1.672 0.719 207.817 1.781 0.828 205.107 1.906 0.953 205.107 2.016 1.063 205.107 2.141 1.188 202.397 2.266 1.313 4.562

Test2 Tensile Strength 286.409 Application Rate 398.3435327 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.828 0 0 0.953 0.125 53.343 1.062 0.234 164.456 1.187 0.359 205.107 1.312 0.484 234.918 1.422 0.594 264.728 1.547 0.719 286.409 1.656 0.828 229.497 1.781 0.953 207.817 1.906 1.078 199.687 2.016 1.188 196.977 2.141 1.313 196.977 2.266 1.438 194.267

Test3 Tensile Strength 302.669 Application Rate 430.5391181 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.828 0 0 0.937 0.109 104.835 1.062 0.234 180.716 1.172 0.344 215.947 1.297 0.469 237.628 1.422 0.594 264.728 1.531 0.703 302.669 1.656 0.828 291.829 1.766 0.938 232.208 1.891 1.063 210.527 2.016 1.188 205.107 2.125 1.297 205.107 2.25 1.422 205.107 2.359 1.531 131.935

Test4 Tensile Strength 221.367 Application Rate 616.6211699 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.828 0 0 0.953 0.125 140.065 1.078 0.25 196.977 1.187 0.359 221.367 1.312 0.484 215.947 1.422 0.594 213.237 1.547 0.719 213.237 1.656 0.828 213.237 1.781 0.953 213.237 1.906 1.078 210.527 2.016 1.188 213.237 2.141 1.313 123.805 2.25 1.422 58.764 2.375 1.547 53.343

CRS65 3%- 2 Hours - Granite SET 7 - Completed 2010 30C Test5 Test6 Test7 Test8 Tensile Strength 454.433 Tensile Strength 443.592 Tensile Strength 438.172 Tensile Strength 435.462 Application Rate 548.83213 Application Rate 617.81616 Application Rate 738.90725 Application Rate 733.10101 Failure Type Adhesion on Steel Failure Type Adhesion on Aggregate Failure Type Adhesion on Steel Failure Type Adhesion on Steel Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.594 0 0 0.344 0 0 0.469 0 0 0.359 0 0 0.719 0.125 278.279 0.469 0.125 134.645 0.594 0.125 286.409 0.484 0.125 178.006 0.828 0.234 408.362 0.594 0.25 356.87 0.703 0.234 402.942 0.594 0.235 365.001 0.953 0.359 438.172 0.703 0.359 419.202 0.828 0.359 427.332 0.719 0.36 416.492 1.078 0.484 449.013 0.828 0.484 438.172 0.937 0.468 432.752 0.828 0.469 427.332 1.187 0.593 451.723 0.937 0.593 440.882 1.062 0.593 438.172 0.953 0.594 435.462 1.312 0.718 451.723 1.062 0.718 443.592 1.187 0.718 112.965 1.078 0.719 367.711 1.422 0.828 454.433 1.172 0.828 37.083 1.547 0.953 451.723 1.672 1.078 37.083

Test1 Tensile Strength 514.054 Application Rate 216.4437895 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.844 0 0 0.969 0.125 129.225 1.094 0.25 188.847 1.203 0.359 218.657 1.328 0.484 226.787 1.437 0.593 229.497 1.562 0.718 229.497 1.687 0.843 232.208 1.797 0.953 234.918 1.922 1.078 240.338 2.031 1.187 245.758 2.156 1.312 256.598 2.281 1.437 270.148 2.391 1.547 280.989 2.516 1.672 297.249 2.625 1.781 316.219 2.75 1.906 346.03 2.859 2.015 378.551 2.984 2.14 421.912 3.109 2.265 467.983 3.219 2.375 514.054 3.344 2.5 500.504 3.453 2.609 316.219 3.578 2.734 237.628 3.703 2.859 213.237 3.812 2.968 205.107 3.937 3.093 118.385

CRS65 3%- 6 Hours - Granite SET D - Completed 2009 25C Test2 Test3 Tensile Strength 888.043 Tensile Strength 985.605 Application Rate 414.7795423 Application Rate 377.7711767 Failure Type Failure Type Time (Sec) Time (Sec) Tensile (kPa) Time (Sec) Time (Sec) Tensile (kPa) 0.937 0 0 0.953 0 0 1.062 0.125 140.065 1.078 0.125 142.775 1.172 0.235 199.687 1.187 0.234 196.977 1.297 0.36 232.208 1.312 0.359 218.657 1.422 0.485 264.728 1.422 0.469 237.628 1.531 0.594 299.959 1.547 0.594 262.018 1.656 0.719 337.9 1.672 0.719 286.409 1.766 0.829 381.261 1.781 0.828 318.93 1.891 0.954 421.912 1.906 0.953 354.16 2.016 1.079 465.273 2.016 1.063 392.101 2.125 1.188 511.344 2.141 1.188 435.462 2.25 1.313 557.415 2.266 1.313 478.823 2.359 1.422 603.486 2.375 1.422 527.604 2.484 1.547 654.977 2.5 1.547 573.675 2.609 1.672 698.338 2.609 1.656 619.747 2.719 1.782 744.409 2.734 1.781 663.108 2.844 1.907 795.901 2.844 1.891 706.469 2.953 2.016 841.972 2.969 2.016 757.96 3.078 2.141 888.043 3.094 2.141 801.321 3.187 2.25 936.824 3.203 2.25 850.102 3.312 2.375 982.895 3.328 2.375 896.173 3.422 2.485 1028.966 3.437 2.484 942.244 3.547 2.61 1080.457 3.562 2.609 985.605 3.672 2.735 1123.818 3.687 2.734 795.901 3.781 2.844 1167.179 3.797 2.844 687.498 3.906 2.969 882.623 3.922 2.969 720.019 4.031 3.094 901.593 4.047 3.094 755.25

CRS65 3% - 6 Hours - Granite SET 7 - Completed 2010 30C Test4 Test5 Test6 Tensile Strength 836.552 Tensile Strength 809.451 Tensile Strength 828.421 Application Rate 1010.3285 Application Rate 977.59783 Application Rate 869.27702 Failure Type Adhesion Failure Type Adhesion on Steel Failure Type Adhesion on Aggregate Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.344 0 0 0.594 0 0 0.469 0 0 0.469 0.125 299.959 0.719 0.125 210.527 0.594 0.125 213.237 0.594 0.25 427.332 0.828 0.234 392.101 0.719 0.25 392.101 0.703 0.359 486.953 0.953 0.359 462.563 0.828 0.359 465.273 0.828 0.484 560.125 1.062 0.468 527.604 0.953 0.484 533.025 0.937 0.593 649.557 1.187 0.593 614.326 1.062 0.593 619.747 1.062 0.718 752.54 1.312 0.718 714.599 1.187 0.718 714.599 1.172 0.828 836.552 1.422 0.828 809.451 1.312 0.843 823.001 1.297 0.953 560.125 1.547 0.953 546.575 1.422 0.953 828.421 1.422 1.078 503.214 1.672 1.078 492.374 1.547 1.078 554.705 1.531 1.187 489.664 1.781 1.187 476.113 1.656 1.187 495.084 1.656 1.312 486.953 1.906 1.312 473.403 1.781 1.312 481.533 1.766 1.422 110.255 2.016 1.422 408.362 1.906 1.437 56.053

Test1 Tensile Strength 809.451 Application Rate 378.2481308 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.594 0 0 0.719 0.125 96.704 0.828 0.234 172.586 0.953 0.359 210.527 1.078 0.484 240.338 1.187 0.593 270.148 1.312 0.718 305.379 1.422 0.828 343.32 1.547 0.953 383.971 1.672 1.078 427.332 1.781 1.187 470.693 1.906 1.312 514.054 2.016 1.422 557.415 2.141 1.547 603.486 2.266 1.672 646.847 2.375 1.781 692.918 2.5 1.906 736.279 2.609 2.015 779.64 2.734 2.14 809.451 2.844 2.25 606.196 2.969 2.375 636.007 3.094 2.5 671.238 3.203 2.609 703.758 3.328 2.734 736.279 3.437 2.843 771.51 3.562 2.968 806.741 3.672 3.078 801.321 3.797 3.203 785.06 3.922 3.328 771.51 4.031 3.437 760.67 4.156 3.562 752.54

CRS65 3% - 24 Hours - Granite SET D - Completed 2009 25C Test2 Tensile Strength 888.043 Application Rate 254.1617478 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.828 0 0 0.953 0.125 115.675 1.062 0.234 178.006 1.187 0.359 207.817 1.312 0.484 215.947 1.422 0.594 215.947 1.547 0.719 218.657 1.656 0.828 218.657 1.781 0.953 218.657 1.906 1.078 221.367 2.016 1.188 226.787 2.141 1.313 234.918 2.25 1.422 251.178 2.375 1.547 275.569 2.5 1.672 310.799 2.609 1.781 348.74 2.734 1.906 389.391 2.844 2.016 435.462 2.969 2.141 478.823 3.078 2.25 524.894 3.203 2.375 570.965 3.328 2.5 617.036 3.437 2.609 663.108 3.562 2.734 706.469 3.672 2.844 755.25 3.797 2.969 798.611 3.922 3.094 847.392 4.031 3.203 888.043 4.156 3.328 747.12 4.266 3.438 687.498 4.391 3.563 725.439


CRS65 3% - 24 Hours - Granite SET 7 - Completed 2010 30C Test4 Test5 Test6 Tensile Strength 806.741 Tensile Strength 1018.126 Tensile Strength 1142.789 Application Rate 974.32488 Application Rate 944.45826 Application Rate 962.754 Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.484 0 0 0.469 0 0 0.469 0 0 0.609 0.125 167.166 0.594 0.125 140.065 0.578 0.109 210.527 0.719 0.235 386.681 0.719 0.25 365.001 0.703 0.234 389.391 0.844 0.36 457.143 0.828 0.359 451.723 0.828 0.359 462.563 0.953 0.469 522.184 0.953 0.484 514.054 0.937 0.468 527.604 1.078 0.594 603.486 1.062 0.593 592.646 1.062 0.593 614.326 1.203 0.719 701.048 1.187 0.718 695.628 1.187 0.718 709.179 1.312 0.828 806.741 1.312 0.843 798.611 1.297 0.828 806.741 1.437 0.953 757.96 1.422 0.953 904.303 1.422 0.953 917.853 1.547 1.063 543.865 1.547 1.078 1018.126 1.531 1.062 1037.096 1.672 1.188 497.794 1.656 1.187 722.729 1.656 1.187 1142.789 1.797 1.313 486.953 1.781 1.312 543.865 1.766 1.297 714.599 1.906 1.422 164.456 1.891 1.422 497.794 1.891 1.422 538.445 2.016 1.547 484.243 2.016 1.547 492.374 2.141 1.672 9.982 2.125 1.656 478.823 2.25 1.781 12.692

CRS65 3% - 2 Hours - Tillite SET D - Completed 2009 25C Test1 Tensile Strength 275.569 Application Rate 463.9208754 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 1.062 0 0 1.187 0.125 91.284 1.297 0.235 178.006 1.422 0.36 215.947 1.531 0.469 248.468 1.656 0.594 275.569 1.781 0.719 226.787 1.891 0.829 207.817 2.016 0.954 202.397 2.125 1.063 199.687 2.25 1.188 199.687 2.359 1.297 183.426

Test2 Tensile Strength 237.628 Application Rate 506.6695096 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.859 0 0 0.969 0.11 64.184 1.094 0.235 167.166 1.219 0.36 207.817 1.328 0.469 237.628 1.453 0.594 218.657 1.562 0.703 205.107 1.687 0.828 199.687 1.797 0.938 199.687 1.922 1.063 199.687 2.047 1.188 199.687 2.156 1.297 18.113

Test3 Tensile Strength 224.077 Application Rate 477.7761194 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.828 0 0 0.953 0.125 77.734 1.062 0.234 175.296 1.187 0.359 205.107 1.297 0.469 224.077 1.422 0.594 205.107 1.547 0.719 202.397 1.656 0.828 199.687 1.781 0.953 196.977 1.891 1.063 196.977 2.016 1.188 196.977 2.125 1.297 18.113

Test4 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.828 0 0.937 0.109 1.062 0.234 1.172 0.344 1.297 0.469 1.422 0.594 1.531 0.703 1.656 0.828 1.766 0.938 1.891 1.063 2.016 1.188 2.125 1.297

275.569 463.9208754 Adhesion Tensile (kPa) 0 104.835 183.426 210.527 240.338 275.569 262.018 213.237 199.687 194.267 188.847 7.272

CRS65 3% - 2 Hours - TILLITE SET 7 - Completed 2010 30C Test5 Test6 Test7 Test8 Tensile Strength 427.332 Tensile Strength 449.013 Tensile Strength 432.752 Tensile Strength 457.143 Application Rate 720.62732 Application Rate 957.3838 Application Rate 899.59915 Application Rate 976.80128 Failure Type Adhesion on Aggregate Failure Type Adhesion on Steel Failure Type Adhesion on Steel Failure Type Adhesion on Steel Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.469 0 0 0.359 0 0 0.484 0 0 0.594 0 0 0.594 0.125 229.497 0.469 0.11 243.048 0.594 0.11 245.758 0.719 0.125 299.959 0.719 0.25 375.841 0.594 0.235 375.841 0.719 0.235 381.261 0.828 0.234 394.811 0.828 0.359 413.782 0.703 0.344 438.172 0.844 0.36 432.752 0.953 0.359 454.433 0.953 0.484 424.622 0.828 0.469 449.013 0.953 0.469 421.912 1.062 0.468 457.143 1.062 0.593 427.332 0.953 0.594 427.332 1.078 0.594 419.202 1.187 0.593 430.042 1.187 0.718 251.178 1.062 0.703 421.912 1.187 0.703 416.492 1.297 0.703 419.202 1.187 0.828 56.053 1.312 1.312 224.077 1.422 0.828 245.758

Test1 Tensile Strength 421.912 Application Rate 359.9931741 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.844 0 0 0.953 0.109 80.444 1.078 0.234 183.426 1.187 0.343 221.367 1.312 0.468 245.758 1.422 0.578 267.438 1.547 0.703 286.409 1.672 0.828 310.799 1.781 0.937 340.61 1.906 1.062 378.551 2.016 1.172 421.912 2.141 1.297 367.711 2.25 1.406 267.438 2.375 1.531 232.208 2.5 1.656 215.947 2.609 1.765 213.237 2.734 1.89 210.527

CRS65 3% - 6 Hours - Tillite SET D - Completed 2009 25C Test2 Tensile Strength 419.202 Application Rate 439.8761805 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.828 0 0 0.953 0.125 53.343 1.062 0.234 180.716 1.187 0.359 224.077 1.297 0.469 253.888 1.422 0.594 291.829 1.547 0.719 332.48 1.656 0.828 373.131 1.781 0.953 419.202 1.891 1.063 378.551 2.016 1.188 275.569 2.141 1.313 234.918 2.25 1.422 221.367 2.375 1.547 215.947 2.484 1.656 213.237 2.609 1.781 213.237 2.734 1.906 213.237

Test3 Tensile Strength 435.462 Application Rate 284.4297845 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.719 0 0 0.828 0.109 104.835 0.937 0.218 188.847 1.062 0.343 224.077 1.187 0.468 237.628 1.297 0.578 234.918 1.422 0.703 234.918 1.547 0.828 253.888 1.656 0.937 278.279 1.781 1.062 305.379 1.891 1.172 337.9 2.016 1.297 367.711 2.125 1.406 402.942 2.25 1.531 435.462 2.375 1.656 383.971 2.484 1.765 375.841 2.609 1.89 367.711

CRS65 3% - 6 Hours - Tillite SET 7 - Completed 2010 30C Test4 Test5 Test6 Tensile Strength 801.321 Tensile Strength 823.001 Tensile Strength 915.143 Application Rate 967.77899 Application Rate 993.96256 Application Rate 960.27597 Failure Type Cohesion Failure Type Adhesion on Steel Failure Type Adhesion on Steel Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.469 0 0 0.719 0 0 0.469 0 0 0.594 0.125 278.279 0.828 0.109 294.539 0.578 0.109 234.918 0.719 0.25 413.782 0.953 0.234 416.492 0.703 0.234 386.681 0.828 0.359 465.273 1.062 0.343 473.403 0.828 0.359 454.433 0.953 0.484 522.184 1.187 0.468 541.155 0.937 0.468 522.184 1.062 0.593 603.486 1.312 0.593 627.877 1.062 0.593 608.906 1.187 0.718 701.048 1.422 0.703 728.149 1.172 0.703 703.758 1.297 0.828 801.321 1.547 0.828 823.001 1.297 0.828 806.741 1.422 0.953 747.12 1.672 0.953 549.285 1.422 0.953 915.143 1.547 1.078 533.025 1.781 1.062 486.953 1.531 1.062 798.611 1.656 1.187 484.243 1.906 1.187 467.983 1.656 1.187 541.155 1.781 1.312 470.693 2.016 1.297 462.563 1.766 1.297 484.243 1.891 1.422 18.113 2.141 1.422 167.166 1.891 1.422 467.983


CRS65 3% - 24 Hours - Tillite SET D - Completed 2009 25C Test2 Tensile Strength 1004.575 Application Rate 404.2555332 Failure Type Time (Sec) Time (Sec) Tensile (kPa) 0.859 0 0 0.969 0.11 102.125 1.094 0.235 188.847 1.203 0.344 224.077 1.328 0.469 253.888 1.437 0.578 291.829 1.562 0.703 332.48 1.687 0.828 375.841 1.797 0.938 416.492 1.922 1.063 457.143 2.031 1.172 505.924 2.156 1.297 549.285 2.281 1.422 600.776 2.391 1.532 644.137 2.516 1.657 687.498 2.625 1.766 733.569 2.75 1.891 779.64 2.875 2.016 825.711 2.984 2.125 874.492 3.109 2.25 917.853 3.219 2.36 961.214 3.344 2.485 1004.575 3.469 2.61 798.611 3.578 2.719 774.22 3.703 2.844 812.161 3.812 2.953 841.972 3.937 3.078 877.203 4.047 3.188 909.723 4.172 3.313 912.433 4.297 3.438 896.173 4.406 3.547 885.333


CRS65 3%- 24 Hours - Tillite SET 7 - Completed 2010 30C Test4 Test5 Test6 Tensile Strength 974.765 Tensile Strength 1156.339 Tensile Strength 1156.339 Application Rate 742.96113 Application Rate 881.35595 Application Rate 881.35595 Failure Type Cohesion Failure Type Cohesion Failure Type Cohesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.469 0 0 0.594 0 0 0.594 0 0 0.594 0.125 47.923 0.719 0.125 42.503 0.719 0.125 28.953 0.703 0.234 332.48 0.828 0.234 321.64 0.828 0.234 313.509 0.828 0.359 419.202 0.953 0.359 416.492 0.953 0.359 416.492 0.937 0.468 443.592 1.078 0.484 481.533 1.078 0.484 473.403 1.062 0.593 457.143 1.187 0.593 551.995 1.187 0.593 549.285 1.187 0.718 492.374 1.312 0.718 641.427 1.312 0.718 636.007 1.297 0.828 570.965 1.422 0.828 736.279 1.422 0.828 733.569 1.422 0.953 660.397 1.547 0.953 841.972 1.547 0.953 836.552 1.531 1.062 760.67 1.656 1.062 953.084 1.672 1.078 942.244 1.656 1.187 860.942 1.781 1.187 1061.487 1.781 1.187 1053.357 1.781 1.312 974.765 1.906 1.312 1156.339 1.906 1.312 1156.339 1.891 1.422 717.309 2.016 1.422 638.717 2.016 1.422 823.001 2.016 1.547 533.025 2.141 1.547 508.634 2.141 1.547 543.865 2.141 1.672 486.953 2.25 1.656 470.693 2.266 1.672 478.823 2.25 1.781 356.87 2.375 2.375 1.781 459.853

Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.594 0 0.703 0.109 0.828 0.234 0.953 0.359 1.062 0.468 1.187 0.593 1.297 0.703 1.422 0.828

Test2 497.794 Tensile Strength 839.450253 Application Rate Adhesion on Aggregate Failure Type Tensile (kPa) Time (Sec) Time (Sec) 0 0.719 0 218.657 0.828 0.109 416.492 0.953 0.234 486.953 1.062 0.343 495.084 1.187 0.468 497.794 1.297 0.578 497.794 1.422 0.703 129.225

SS60 - 2 Hours - Granite SET 8 - Completed 2010 30C Test3 Test4 476.113 Tensile Strength 473.403 Tensile Strength 823.7249135 Application Rate 798.3187184 Application Rate Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Tensile (kPa) Time (Sec) Time (Sec) Tensile (kPa) Time (Sec) Time (Sec) 0 0.594 0 0 0.719 0 259.308 0.703 0.109 286.409 0.844 0.125 421.912 0.828 0.234 421.912 0.953 0.234 462.563 0.953 0.359 457.143 1.078 0.359 473.403 1.062 0.468 470.693 1.187 0.468 476.113 1.187 0.593 473.403 1.312 0.593 80.444 1.297 0.703 473.403 1.437 0.718 1.422 0.828 64.184

Test5 459.853 Tensile Strength 775.4688027 Application Rate Adhesion on Aggregate Failure Type Tensile (kPa) Time (Sec) Time (Sec) 0 0.594 0 308.089 0.703 0.109 421.912 0.828 0.234 451.723 0.953 0.359 457.143 1.062 0.468 459.853 1.187 0.593 39.793 1.312 0.718

478.823 766.3288364 Adhesion on Aggregate Tensile (kPa) 0 318.93 421.912 478.823 459.853 454.433 123.805

SS60 - 2 Hours - Granite SET 9 - Completed 2010 30C Test6 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.703 0 0.812 0.109 0.937 0.234 1.062 0.359 1.172 0.469 1.297 0.594 1.406 0.703 1.531 0.828

Test7 432.752 Tensile Strength 615.5789474 Application Rate Adhesion on Aggregate Failure Type Tensile (kPa) Time (Sec) Time (Sec) 0 0.703 0 186.136 0.828 0.125 362.291 0.953 0.25 411.072 1.062 0.359 427.332 1.187 0.484 430.042 1.297 0.594 432.752 1.422 0.719 69.604 1.531 0.828

419.202 583.0347705 Adhesion on Aggregate and Steel Tensile (kPa) 0 275.569 378.551 405.652 416.492 419.202 419.202 39.793

Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.594 0 0.703 0.109 0.828 0.234 0.953 0.359 1.062 0.468 1.187 0.593 1.297 0.703 1.422 0.828 1.531 0.937 1.656 1.062

SS65 - 6 Hours - Granite SET 8 - Completed 2010 30C Test2 Test3 470.693 Tensile Strength 467.983 Tensile Strength 784.608769 Application Rate 789.1787521 Application Rate Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Tensile (kPa) Time (Sec) Time (Sec) Tensile (kPa) Time (Sec) Time (Sec) 0 0.719 0 0 0.578 0 85.864 0.828 0.109 234.918 0.703 0.125 359.58 0.953 0.234 405.652 0.828 0.25 438.172 1.062 0.343 451.723 0.937 0.359 459.853 1.187 0.468 465.273 1.062 0.484 465.273 1.312 0.593 467.983 1.172 0.594 470.693 1.422 0.703 454.433 1.297 0.719 467.983 1.406 0.828 467.983 1.531 0.953 80.444

459.853 774.1632997 Adhesion on Aggregate Tensile (kPa) 0 175.296 381.261 457.143 459.853 459.853 459.853 457.143 191.557

SS60 - 6 Hours - Granite SET 9 - Completed 2010 30C Test4 Test5 Test6 Tensile Strength 467.983 Tensile Strength 465.273 Tensile Strength 457.143 Application Rate 650.88039 Application Rate 661.83926 Application Rate 636.68942 Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.828 0 0 0.719 0 0 0.719 0 0 0.953 0.125 93.994 0.828 0.109 183.426 0.844 0.125 210.527 1.062 0.234 356.87 0.953 0.234 389.391 0.953 0.234 389.391 1.187 0.359 438.172 1.078 0.359 454.433 1.078 0.359 443.592 1.312 0.484 457.143 1.187 0.468 462.563 1.187 0.468 451.723 1.422 0.594 465.273 1.312 0.593 462.563 1.312 0.593 457.143 1.547 0.719 467.983 1.422 0.703 465.273 1.437 0.718 457.143 1.656 0.828 467.983 1.547 0.828 462.563 1.547 0.828 457.143 1.781 0.953 467.983 1.672 0.953 386.681 1.672 0.953 457.143 1.891 1.063 467.983 1.781 1.062 194.267 2.016 1.188 77.734

SS60 - 24 Hours - Granite SET 8 - Completed 2010 30C Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.828 0 0.937 0.109 1.062 0.234 1.187 0.359 1.297 0.469 1.422 0.594 1.531 0.703 1.656 0.828 1.781 0.953 1.891 1.063

Test2 495.084 Tensile Strength 696.5348506 Application Rate Adhesion on Aggregate Failure Type Tensile (kPa) Time (Sec) Time (Sec) 0 0.937 0 283.699 1.062 0.125 424.622 1.172 0.235 489.664 1.297 0.36 495.084 1.422 0.485 492.374 1.531 0.594 489.664 1.656 0.719 489.664 1.766 0.829 489.664 1.891 0.954 23.533 2.016 1.079 2.125 1.188 2.25 1.313 2.359 1.422 2.484 1.547

473.403 658.418637 Adhesion on Aggregate Tensile (kPa) 0 313.509 427.332 459.853 467.983 473.403 473.403 481.533 500.504 503.214 484.243 481.533 481.533 102.125

SS60 - 24 Hours - Granite SET 9 - Completed 2010 30C Test3 Test4 Test5 Test6 Test7 Tensile Strength 668.528 Tensile Strength 462.563 Tensile Strength 467.983 Tensile Strength 427.332 Tensile Strength 478.823 Application Rate 715.75604 Application Rate 729.76728 Application Rate 789.17875 Application Rate 705.72727 Application Rate 728.53872 Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.719 0 0 0.594 0 0 0.469 0 0 0.828 0 0 0.953 0 0 0.844 0.125 183.426 0.703 0.109 83.154 0.594 0.125 156.326 0.937 0.109 102.125 1.062 0.109 291.829 0.953 0.234 362.291 0.828 0.234 329.77 0.703 0.234 351.45 1.062 0.234 329.77 1.187 0.234 392.101 1.078 0.359 424.622 0.937 0.343 411.072 0.828 0.359 416.492 1.172 0.344 405.652 1.312 0.359 446.303 1.187 0.468 495.084 1.062 0.468 462.563 0.937 0.468 459.853 1.297 0.469 427.332 1.422 0.469 478.823 1.312 0.593 576.386 1.187 0.593 432.752 1.062 0.593 467.983 1.422 0.594 419.202 1.547 0.594 432.752 1.437 0.718 668.528 1.297 0.703 427.332 1.172 0.703 432.752 1.531 0.703 419.202 1.656 0.703 421.912 1.547 0.828 592.646 1.422 0.828 424.622 1.297 0.828 424.622 1.656 0.828 419.202 1.781 0.828 416.492 1.672 0.953 478.823 1.531 0.937 424.622 1.406 0.937 421.912 1.766 0.938 416.492 1.906 0.953 58.764 1.781 1.062 446.303 1.656 1.062 169.876 1.531 1.062 421.912 1.891 1.063 416.492

Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.594 0 0.703 0.109 0.828 0.234 0.937 0.343 1.062 0.468 1.172 0.578

Test2 443.592 Tensile Strength 947.8461538 Application Rate Adhesion on Aggregate Failure Type Tensile (kPa) Time (Sec) Time (Sec) 0 0.703 0 302.669 0.828 0.125 405.652 0.953 0.25 432.752 1.062 0.359 443.592 1.187 0.484 443.592 1.297 0.594 1.422 0.719

SS60 - 2 Hours - Tillite SET 8 - Completed 2010 30C Test3 Test4 Test5 438.172 Tensile Strength 430.042 Tensile Strength 454.433 Tensile Strength 413.782 905.3140496 Application Rate 913.10256 Application Rate 971.01068 Application Rate 884.14957 Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Tensile (kPa) Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0 0.594 0 0 0.594 0 0 0.719 0 0 302.669 0.703 0.109 183.426 0.703 0.109 267.438 0.828 0.109 112.965 402.942 0.828 0.234 362.291 0.828 0.234 386.681 0.953 0.234 335.19 430.042 0.937 0.343 416.492 0.937 0.343 440.882 1.062 0.343 408.362 438.172 1.062 0.468 427.332 1.062 0.468 454.433 1.187 0.468 413.782 438.172 1.187 0.593 430.042 1.172 0.578 435.462 1.312 0.593 413.782 131.935 1.297 0.703 413.782 1.297 0.703 430.042 1.422 0.703 413.782 1.422 0.828 427.332 1.547 0.828 413.782 1.531 0.937 1.852 1.656 0.937 64.184

SS60 - 2 Hours - Tillite SET 9 - Completed 2010 30C Test6 Test7 Tensile Strength 416.492 Tensile Strength 402.942 Application Rate 701.16498 Application Rate 678.35354 Failure Type Adhesion on Aggregate Failure Type Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.953 0 0 0.703 0 0 1.062 0.109 270.148 0.828 0.125 175.296 1.187 0.234 373.131 0.937 0.234 346.03 1.297 0.344 400.231 1.062 0.359 386.681 1.422 0.469 411.072 1.187 0.484 397.521 1.547 0.594 416.492 1.297 0.594 402.942 1.656 0.703 413.782 1.422 0.719 402.942 1.781 0.828 18.113 1.531 0.828 99.414

Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.719 0 0.828 0.109 0.953 0.234 1.062 0.343 1.187 0.468 1.312 0.593 1.422 0.703 1.547 0.828

SS60- 6 Hours - Tillite SET 8 - Completed 2010 30C Test2 Test3 457.143 Tensile Strength 451.723 Tensile Strength 770.8988196 Application Rate 760.476431 Application Rate Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Tensile (kPa) Time (Sec) Time (Sec) Tensile (kPa) Time (Sec) Time (Sec) 0 0.828 0 0 0.703 0 321.64 0.953 0.125 340.61 0.828 0.125 421.912 1.062 0.234 419.202 0.937 0.234 451.723 1.187 0.359 440.882 1.062 0.359 454.433 1.297 0.469 446.303 1.172 0.469 457.143 1.422 0.594 451.723 1.297 0.594 457.143 1.547 0.719 449.013 1.422 0.719 188.847 1.656 0.828 449.013 1.531 0.828

470.693 755.9141414 Adhesion on Aggregate Tensile (kPa) 0 302.669 408.362 470.693 467.983 449.013 440.882 440.882

SS60- 6 Hours - Tillite SET 9 - Completed 2010 30C Test4 Test5 Test6 Tensile Strength 470.693 Tensile Strength 551.995 Tensile Strength 587.226 Application Rate 766.32884 Application Rate 955.00865 Application Rate 988.59596 Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.719 0 0 0.828 0 0 0.828 0 0 0.828 0.109 150.906 0.937 0.109 305.379 0.953 0.125 270.148 0.953 0.234 367.711 1.062 0.234 411.072 1.078 0.25 394.811 1.078 0.359 443.592 1.172 0.344 473.403 1.187 0.359 446.303 1.187 0.468 470.693 1.297 0.469 551.995 1.312 0.484 505.924 1.312 0.593 454.433 1.406 0.578 551.995 1.422 0.594 587.226 1.422 0.703 451.723 1.531 0.703 473.403 1.547 0.719 484.243 1.547 0.828 449.013 1.656 0.828 454.433 1.656 0.828 454.433 1.656 0.937 451.723 1.766 0.938 449.013 1.781 0.953 446.303 1.781 1.062 408.362 1.891 1.063 169.876 1.906 1.078 39.793

SS60 - 24 Hours - Tillite SET 8 - Completed 2010 30C Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.828 0 0.937 0.109 1.062 0.234 1.172 0.344 1.297 0.469 1.422 0.594 1.531 0.703 1.656 0.828 1.766 0.938 1.891 1.063 2.016 1.188 2.125 1.297

Test2 755.25 Tensile Strength 702.6642512 Application Rate Adhesion on Aggregate Failure Type Tensile (kPa) Time (Sec) Time (Sec) 0 0.937 0 286.409 1.062 0.125 424.622 1.187 0.25 489.664 1.297 0.36 562.835 1.422 0.485 649.557 1.531 0.594 755.25 1.656 0.719 581.806 1.781 0.844 511.344 1.891 0.954 492.374 2.016 1.079 489.664 15.402

598.066 692.3421419 Adhesion on Aggregate Tensile (kPa) 0 289.119 421.912 484.243 557.415 598.066 497.794 478.823 473.403 467.983

SS60 - 24 Hours - Tillite SET 9 - Completed 2010 30C Test3 Test4 Test5 Test6 Test7 Tensile Strength 912.433 Tensile Strength 524.894 Tensile Strength 625.167 Tensile Strength 554.705 Tensile Strength 576.386 Application Rate 846.4128 Application Rate 883.65993 Application Rate 869.49513 Application Rate 933.8468 Application Rate 810.12603 Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type Adhesion on Aggregate Failure Type & Steel Adhesion on Aggregate Failure Type Adhesion on Aggregate Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.828 0 0 1.062 0 0 0.703 0 0 1.187 0 0 0.828 0 0 0.953 0.125 226.787 1.187 0.125 102.125 0.828 0.125 215.947 1.297 0.11 240.338 0.953 0.125 286.409 1.078 0.25 367.711 1.297 0.235 310.799 0.953 0.25 354.16 1.422 0.235 356.87 1.078 0.25 359.58 1.187 0.359 419.202 1.422 0.36 392.101 1.062 0.359 405.652 1.531 0.344 411.072 1.187 0.359 383.971 1.312 0.484 465.273 1.547 0.485 451.723 1.187 0.484 462.563 1.656 0.469 478.823 1.312 0.484 392.101 1.437 0.609 533.025 1.656 0.594 524.894 1.297 0.594 541.155 1.781 0.594 554.705 1.422 0.594 400.231 1.547 0.719 619.747 1.781 0.719 438.172 1.422 0.719 625.167 1.891 0.704 459.853 1.547 0.719 430.042 1.672 0.844 714.599 1.891 0.829 413.782 1.547 0.844 543.865 2.016 0.829 408.362 1.672 0.844 492.374 1.781 0.953 817.581 2.016 0.954 405.652 1.656 0.953 443.592 2.125 0.938 397.521 1.781 0.953 576.386 1.906 1.078 912.433 2.125 1.063 402.942 1.781 1.078 416.492 2.25 1.063 392.101 1.906 1.078 541.155 2.016 1.188 562.835 2.25 1.188 402.942 1.891 1.188 329.77 2.359 1.172 88.574 2.016 1.188 430.042 2.141 1.313 459.853 2.375 1.313 110.255 2.141 1.313 405.652 2.266 1.438 432.752 2.25 1.422 397.521 2.375 1.547 427.332 2.375 1.547 88.574

SS60 +3% Latex- 2 Hours - Granite SET 10 - Completed 2010 30C Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 1.062 0 1.172 0.11 1.297 0.235 1.422 0.36 1.531 0.469 1.656 0.594 1.766 0.704 1.891 0.829

462.563 657.0497159 Adhesion Tensile (kPa) 0 310.799 416.492 446.303 457.143 459.853 462.563 362.291

Test2 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.828 0 0.953 0.125 1.062 0.234 1.187 0.359 1.312 0.484 1.422 0.594 1.547 0.719 1.672 0.844 1.781 0.953

457.143 635.8038943 Adhesion Tensile (kPa) 0 153.616 359.58 432.752 449.013 454.433 457.143 457.143 145.486




449.013 746.7878788 Adhesion Tensile (kPa) 0 205.107 378.551 430.042 440.882 443.592 446.303 449.013 446.303 23.533



SS60+3% Latex - 2 Hours - Granite SET 12 - Completed 2010 30C Test6 Tensile Strength 473.403 Application Rate 673.4039829 Failure Type Adhesion Time (Sec) Time (Sec) Tensile (kPa) 0.953 0 0 1.062 0.109 327.06 1.187 0.234 430.042 1.297 0.344 459.853 1.422 0.469 470.693 1.547 0.594 473.403 1.656 0.703 473.403 1.781 0.828 473.403 1.891 0.938 83.154



Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 1.062 0 1.187 0.125 1.297 0.235 1.422 0.36 1.547 0.485 1.656 0.594 1.781 0.719 1.891 0.829 2.016 0.954 2.141 1.079 2.25 1.188

497.794 688.5730181 Adhesion Tensile (kPa) 0 88.574 370.421 459.853 486.953 492.374 495.084 497.794 495.084 495.084 495.084

SS60+3% Latex - 6 Hours - Granite SET 10 - Completed 2010 30C Test2 Test3 Tensile Strength 492.374 Tensile Strength Application Rate 684.8038943 Application Rate Failure Type Adhesion Failure Type Time (Sec) Time (Sec) Tensile (kPa) Time (Sec) Time (Sec) 1.062 0 0 1.062 0 1.187 0.125 134.645 1.187 0.125 1.297 0.235 383.971 1.312 0.25 1.422 0.36 459.853 1.422 0.36 1.531 0.469 481.533 1.547 0.485 1.656 0.594 489.664 1.672 0.61 1.781 0.719 492.374 1.781 0.719 1.891 0.829 492.374 1.906 0.844 2.016 0.954 492.374 2.016 0.954 2.125 1.063 492.374 2.141 1.079 2.25 1.188 83.154 2.266 1.204

486.953 673.4951321 Adhesion Tensile (kPa) 0 159.036 389.391 454.433 476.113 481.533 484.243 484.243 486.953 486.953 83.154

SS60+3% LATEX - 6 Hours - Granite SET 12 - Completed 2010 30C Test4 Test5 Test6 Tensile Strength 519.474 Tensile Strength 503.214 Tensile Strength 495.084 Application Rate 627.38406 Application Rate 607.74638 Application Rate 594.65459 Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.844 0 0 0.703 0 0 0.828 0 0 0.969 0.125 194.267 0.828 0.125 237.628 0.953 0.125 302.669 1.078 0.234 421.912 0.937 0.234 419.202 1.062 0.234 440.882 1.203 0.359 486.953 1.062 0.359 476.113 1.187 0.359 476.113 1.312 0.468 508.634 1.172 0.469 495.084 1.312 0.484 489.664 1.437 0.593 514.054 1.297 0.594 500.504 1.422 0.594 492.374 1.547 0.703 516.764 1.422 0.719 503.214 1.547 0.719 495.084 1.672 0.828 519.474 1.531 0.828 503.214 1.656 0.828 492.374 1.797 0.953 519.474 1.656 0.953 503.214 1.781 0.953 492.374 1.906 1.062 169.876 1.766 1.063 53.343 1.891 1.063 42.503

Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.859 0 0.984 0.125 1.094 0.235 1.219 0.36 1.344 0.485 1.453 0.594 1.578 0.719 1.687 0.828 1.812 0.953 1.937 1.078 2.047 1.188 2.172 1.313

SS60+3% Latex - 24 Hours - Granite SET 10 - Completed 2010 30C Test2 Test3 554.705 Tensile Strength 587.226 Tensile Strength 481.533 933.8468013 Application Rate 944.8789986 Application Rate 1026.72281 Adhesion Failure Type Adhesion Failure Type Adhesion Tensile (kPa) Time (Sec) Time (Sec) Tensile (kPa) Time (Sec) Time (Sec)Tensile (kPa) 0 0.828 0 0 1.078 0 0 104.835 0.937 0.109 234.918 1.187 0.109 183.426 375.841 1.062 0.234 405.652 1.312 0.234 378.551 465.273 1.187 0.359 462.563 1.437 0.359 457.143 522.184 1.297 0.469 511.344 1.547 0.469 481.533 554.705 1.422 0.594 587.226 1.672 0.594 470.693 514.054 1.547 0.719 679.368 1.797 0.719 467.983 505.924 1.656 0.828 562.835 1.906 0.828 465.273 503.214 1.781 0.953 503.214 2.031 0.953 465.273 503.214 1.891 1.063 486.953 2.141 1.063 467.983 503.214 2.016 1.188 484.243 2.266 1.188 467.983 497.794 2.141 1.313 481.533 2.375 1.297 467.983 2.25 1.422 9.982 2.5 1.422 451.723

SS60+ 3% Latex - 24 Hours - Granite SET 12 - Completed 2010 30C Test4 Test5 Test6 Test7 Tensile Strength 476.113 Tensile Strength 484.243 Tensile Strength 473.403 Tensile Strength 530.314 Application Rate 1015.1663 Application Rate 1000.5021 Application Rate 965.22009 Application Rate 1040.4979 Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.828 0 0 0.828 0 0 0.844 0 0 0.594 0 0 0.937 0.109 280.989 0.953 0.125 205.107 0.953 0.109 205.107 0.719 0.125 83.154 1.062 0.234 416.492 1.062 0.234 386.681 1.078 0.234 381.261 0.828 0.234 335.19 1.187 0.359 473.403 1.187 0.359 459.853 1.187 0.343 432.752 0.953 0.359 427.332 1.297 0.469 476.113 1.312 0.484 484.243 1.312 0.468 451.723 1.062 0.468 486.953 1.422 0.594 473.403 1.422 0.594 470.693 1.437 0.593 473.403 1.187 0.593 530.314 1.547 0.719 473.403 1.547 0.719 465.273 1.547 0.703 470.693 1.312 0.718 470.693 1.656 0.828 473.403 1.656 0.828 467.983 1.672 0.828 457.143 1.422 0.828 457.143 1.781 0.953 473.403 1.781 0.953 465.273 1.781 0.937 457.143 1.547 0.953 457.143 1.891 1.063 473.403 1.906 1.078 465.273 1.906 1.062 457.143 1.672 1.078 454.433 2.016 1.188 473.403 2.016 1.188 167.166 2.031 1.187 457.143 1.781 1.187 454.433 2.141 1.313 473.403 2.141 1.297 457.143 1.906 1.312 375.841 2.25 1.422 473.403 2.266 1.422 245.758 2.375 1.547 142.775


438.172 737.6632997 Adhesion Tensile (kPa) 0 77.734 327.06 411.072 430.042 438.172 438.172 438.172 0

SS60+3% LATEX - 2 Hours - Tillite SET 8 - Completed 2010 30C Test2 Test3 Test4 Tensile Strength 438.172 Tensile Strength 432.752 Tensile Strength 419.202 Application Rate 733.1010101 Application Rate 728.53872 Application Rate 701.16498 Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Time (Sec) Time (Sec) Tensile (kPa) Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.953 0 0 1.187 0 0 0.828 0 0 1.062 0.109 150.906 1.297 0.11 240.338 0.937 0.109 297.249 1.187 0.234 354.16 1.422 0.235 378.551 1.062 0.234 394.811 1.297 0.344 416.492 1.531 0.344 424.622 1.172 0.344 413.782 1.422 0.469 430.042 1.656 0.469 432.752 1.297 0.469 416.492 1.547 0.594 435.462 1.781 0.594 432.752 1.422 0.594 416.492 1.656 0.703 435.462 1.891 0.704 432.752 1.531 0.703 419.202 1.781 0.828 438.172 2.016 0.829 432.752 1.656 0.828 419.202 1.891 0.938 416.492 2.125 0.938 432.752 1.766 0.938 419.202 2.25 1.063 432.752 1.891 1.063 110.255 2.359 1.172 39.793

SS60+3% LATEX - 2 Hours - Tillite SET12 - Completed 2010 30C Test6 Test7 Test8 Tensile Strength 457.143 Tensile Strength 449.013 Tensile Strength 443.592 Application Rate 755.91414 Application Rate 748.04722 Application Rate 748.04722 Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.703 0 0 0.594 0 0 0.594 0 0 0.828 0.125 47.923 0.719 0.125 34.373 0.719 0.125 262.018 0.937 0.234 310.799 0.828 0.234 310.799 0.828 0.234 392.101 1.062 0.359 416.492 0.953 0.359 413.782 0.953 0.359 430.042 1.172 0.469 440.882 1.078 0.484 435.462 1.078 0.484 438.172 1.297 0.594 449.013 1.187 0.593 443.592 1.187 0.593 443.592 1.422 0.719 451.723 1.312 0.718 449.013 1.312 0.718 443.592 1.531 0.828 457.143 1.422 0.828 449.013 1.422 0.828 419.202 1.656 0.953 454.433 1.547 0.953 451.723 1.766 1.063 4.562 1.656 1.062 64.184



SS60+3% LATEX - 6 Hours - Tillite SET 8 - Completed 2010 30C Test2 Tensile Strength 467.983 Application Rate 787.8501684 Failure Type Adhesion Time (Sec) Time (Sec) Tensile (kPa) 0.953 0 0 1.062 0.109 280.989 1.187 0.234 416.492 1.312 0.359 476.113 1.422 0.469 467.983 1.547 0.594 467.983 1.672 0.719 465.273 1.781 0.828 467.983 1.906 0.953 467.983 2.016 1.063 131.935



SS60+ 3% Latex 6 Hours - Tillite SET 12 - Completed 2010 30C Test4 Test5 Test6 Tensile Strength 486.953 Tensile Strength 484.243 Tensile Strength 481.533 Application Rate 816.59865 Application Rate 806.09933 Application Rate 798.31872 Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.844 0 0 0.828 0 0 0.594 0 0 0.953 0.109 53.343 0.953 0.125 53.343 0.719 0.125 47.923 1.078 0.234 346.03 1.062 0.234 343.32 0.828 0.234 340.61 1.187 0.343 443.592 1.187 0.359 440.882 0.953 0.359 432.752 1.312 0.468 473.403 1.297 0.469 467.983 1.062 0.468 465.273 1.437 0.593 484.243 1.422 0.594 478.823 1.187 0.593 473.403 1.547 0.703 486.953 1.547 0.719 481.533 1.312 0.718 478.823 1.672 0.828 486.953 1.656 0.828 484.243 1.422 0.828 481.533 1.781 0.937 486.953 1.781 0.953 484.243 1.547 0.953 481.533 1.906 1.062 411.072 1.906 1.078 45.213

SS60+3% LATEX - 24 Hours - Tillite SET 8 - Completed 2010 30C Test1 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.828 0 0.953 0.125 1.078 0.25 1.187 0.359 1.312 0.484 1.422 0.594 1.547 0.719 1.656 0.828 1.781 0.953 1.906 1.078 2.016 1.188 2.141 1.313 2.266 1.438 2.375 1.547 2.5 1.672 2.609 1.781 2.734 1.906

1107.558 932.2878788 Adhesion Tensile (kPa) 0 118.385 351.45 440.882 505.924 587.226 684.788 785.06 893.463 999.155 1107.558 630.587 522.184 492.374 484.243 481.533 356.87

Test2 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 1.062 0 1.187 0.125 1.297 0.235 1.422 0.36 1.531 0.469 1.656 0.594 1.781 0.719 1.891 0.829 2.016 0.954 2.125 1.063 2.25 1.188 2.375 1.313 2.484 1.422 2.609 1.547 2.719 1.657 2.844 1.782


Test3 Tensile Strength Application Rate Failure Type Time (Sec) Time (Sec) 0.953 0 1.062 0.109 1.187 0.234 1.297 0.344 1.422 0.469 1.547 0.594 1.656 0.703 1.781 0.828 1.891 0.938 2.016 1.063 2.141 1.188 2.25 1.297 2.375 1.422 2.484 1.531

649.557 923.9786629 Adhesion Tensile (kPa) 0 232.208 383.971 440.882 489.664 560.125 649.557 554.705 481.533 462.563 457.143 457.143 454.433 47.923

SS60+3% LATEX - 24 Hours - Tillite SET 9 - Completed 2010 30C Test3 Test4 Test5 Test6 Tensile Strength 757.96 Tensile Strength 676.658 Tensile Strength 776.93 Tensile Strength 698.338 Application Rate 839.45025 Application Rate 941.10987 Application Rate 966.38407 Application Rate 988.59596 Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Failure Type Adhesion Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa)Time (Sec) Time (Sec)Tensile (kPa) 0.719 0 0 0.578 0 0 0.719 0 0 0.703 0 0 0.844 0.125 15.402 0.703 0.125 161.746 0.828 0.109 294.539 0.828 0.125 262.018 0.953 0.234 308.089 0.828 0.25 365.001 0.953 0.234 400.231 0.937 0.234 397.521 1.078 0.359 411.072 0.937 0.359 438.172 1.062 0.343 446.303 1.062 0.359 449.013 1.187 0.468 454.433 1.062 0.484 505.924 1.187 0.468 505.924 1.187 0.484 508.634 1.312 0.593 497.794 1.172 0.594 584.516 1.312 0.593 587.226 1.297 0.594 587.226 1.422 0.703 570.965 1.297 0.719 676.658 1.422 0.703 679.368 1.422 0.719 682.078 1.547 0.828 657.687 1.406 0.828 657.687 1.547 0.828 776.93 1.531 0.828 698.338 1.672 0.953 757.96 1.531 0.953 500.504 1.656 0.937 608.906 1.656 0.953 511.344 1.781 1.062 533.025 1.656 1.078 462.563 1.781 1.062 486.953 1.766 1.063 462.563 1.906 1.187 476.113 1.766 1.188 454.433 1.891 1.172 459.853 1.891 1.188 451.723 2.016 1.297 459.853 1.891 1.313 451.723 2.016 1.297 451.723 2.016 1.313 449.013 2.141 1.422 457.143 2.141 1.422 348.74 2.125 1.422 449.013 2.266 1.547 457.143 2.25 1.547 449.013 2.375 1.656 457.143


ADDENDUM 5- PAPER PRESENTED AT 2ND INTERNATIONAL SPRAY SEALING CONFERENCE, AUCKLAND, AUSTRALIA, 2010

123 AH GREYLING

December 2012

nd

2 International Sprayed Sealing Conference – Sustaining sprayed sealing practice, Melbourne, Australia 2010

DEVELOPMENT OF A TEST METHOD FOR DETERMINING EMULSION BOND STRENGTH USING THE BITUMEN BOND STRENGTH (BBS) TEST Timothy Miller, University Wisconsin –Madison, United States of America Andre Greyling, University of Stellenbosch, South Africa Prof. Hussain Bahia, University of Wisconsin – Madison, United States of America Prof. Kim Jenkins, University of Stellenbosch, South Africa

ABSTRACT Quantifying bond strength between bituminous binders and aggregates demands simple, inexpensive test methods as the road building industry moves toward performance-based specifications for surface seals. This research identifies the Pneumatic Adhesive Tensile Testing Instrument, commonly known as PATTI and originally developed for use in the painting industry, as an appropriate instrument for evaluating bond strength development in the newly proposed Bitumen Bond Strength (BBS) test. Recently developed at the University of Wisconsin – Madison in conjunction with the University of Ancona – Italy and the University of Stellenbosch – South Africa, the BBS test procedure has been used successfully to characterize moisture sensitivity and bond strength development in hot-applied binders and bitumen emulsions, respectively [1, 2, 3, 4]. This paper describes the test method and evaluates the relevance of the procedure using bitumen and emulsions produced in U.S.A. and South Africa.

INTRODUCTION Bituminous surface seal design has experienced significant development and enhancement in several countries over the past two decades, particularly in South Africa, Australia, New Zealand and North America. Despite growing surface seal use, selection of binder type and grade does not always follow rigorous scientific evaluation, often relying instead on common practice and best judgment. Historically, empirical information and contractor experience strongly influence surface seal design. With increasing traffic volumes and heightened performance demands, performance-based specifications are being developed to account for deficiencies in surface seal design methodologies [5, 6]. Performance-based specifications for surface seals identify the need for a simple and inexpensive technique for evaluating emulsion bond strength development over time as well as binder-aggregate compatibility. Although various tests exist for investigating adhesion between bituminous emulsions and aggregate chips [7], some of these tests have limited application to performance-based specifications. The Bitumen Bond Strength (BBS) test method aims to address some of the limitations encountered in evaluating bond strength. Researchers at the University of Wisconsin – Madison (UWM) recently developed the BBS test in partnership with the University of Ancona – Italy (UAI) and the University of Stellenbosch – South Africa (USSA) for hot applied binders and emulsions, respectively. USSA became involved in BBS test efforts in 2008 to assist in the evaluation of the BBS test method, while UAI contributed significantly to the development of the test apparatus over the past three years. Developers of the BBS test envision that the test method will be able to quantify and characterize bond strength development, adhesive properties and aggregate-binder compatibility, thereby improving the efficacy of surface seal design. Further research is needed to identify other parameters, such as traffic-related loading rates, in order to more accurately simulate field distresses encountered in early-life phases of surface seals. This paper focuses on characterizing emulsions commonly applied in surface seals. Bond strength development is critical for new surface seals as loose aggregates must be minimized once a road enters into service following construction. Bond strength development depends on several factors, including the material characteristics of the emulsion, environmental conditions during construction, and the mineralogical characteristics of the aggregate chips [8, 9]. The rate at which bond strength develops directly affects an 1

nd


acceptable time threshold at which the seal may be opened to traffic. The seal must have sufficient bond strength to retain aggregate chips, but the seal must also gain this strength in a sufficiently short period of time to minimize user delay. Current evaluation and validation practices for certifying newly applied chip seals often rely on subjective judgment, though various specification systems are under development. Because bond strength development between emulsions and aggregate chips in early-life phases has implications for early raveling, developing a standard test method that evaluates bond strength is critical for reducing premature seal failures.

BBS TEST DEVELOPMENT BACKGROUND Developing practical and effective test methods to characterize bond strength is critical for pavement designers and contractors to predict how the surface seal will perform over the expected service life. Pull-off tests developed for use in other industries can be adapted for evaluation of pavement materials. The painting industry initially developed pull-off tests to evaluate the pull-off strength of coatings on rigid substrates such as metal, concrete and wood. The goal of such pull-off test methods is to measure the maximum normal force that a solid surface coating can withstand before the adhesive detaches from the surface at failure. Such tests allow for the evaluation of failure type (e.g. adhesive or cohesive) through inspection of the failure surface after detachment has occurred [2, 10]. The bitumen industry first utilized the Pneumatic Adhesive Tensile Testing Instrument, or PATTI, in the late 1990s to evaluate adhesive loss of binder-aggregate systems exposed to moisture conditioning [11]. Recent research at UWM continues to investigate the effects of moisture damage for hot applied binders [2, 4, 12]. Initial tests identified several significant effects, including variations in preparing the test assembly (operator dependence), binder film thickness, and curing and testing temperatures. Recent generations of the PATTI, notably the PATTI Quantum Gold (PQG), address these shortcomings while ensuring compliance with surface seal industry requirements [1]. Early generations of PATTI consisted of a pressure hose, adhesion tester, piston, reaction plate and a metal pull-out stub. Figure 1 below shows a typical assembly.

Figure 1 – Original PATTI assembly. The original test procedure entailed the following steps:

2

•

A hot bitumen sample is applied to a glass substrate and allowed to cure for a fixed time interval.

•

A metal pull-out stub is applied to the bitumen sample and allowed to set for a given time interval.

•

After placing the piston over the pull-out stub, the reaction plate is fixed to the stub.

nd


•

The pressure hose introduces compressed air to the piston, resulting in an upward force on the specimen and eventual failure of the binder. Failure occurs when the applied pressure exceeds the cohesive strength of the binder or the adhesive strength of the binder-aggregate interface.

•

The pressure at failure is recorded and the procedure is repeated for other test specimens.

While the original test procedure did yield quantitative information related to bond strength characteristics and failure behavior, research at UWM and UAI identified several factors influencing the efficacy of the test method. Some of these factors include: •

The binder film thickness between the stub and substrate could not be controlled easily.

•

Because the original pull-out stub measured only 12.7 mm in diameter, the stub geometry limited the measurement of smaller tensile strengths. Recent modifications to the pull-out stub design at UWM and UAI improved the geometry by nearly doubling the stub diameter.

•

The device did not report pressure over time, making the calculation of loading rate difficult. Rather, PATTI reported only the real-time applied load but not within a computer-based graphical user interface.

•

The loading rate varied and could not be set easily. While the PATTI is equipped with a rate control dial, the dial did not effectively control the loading rate or report the real-time loading rate.

•

Initial tests were performed on glass substrates, hardly a suitable surface seal material.

INITIAL TEST METHOD LIMITATIONS & MODIFICATIONS Improvements to the pull-out stub design, loading rate control and substrate preparation procedures represent significant advancements in the BBS test method. Film thickness is effectively controlled with an improved stub design. Loading rate is effectively controlled with the PQG. Substrate surface characteristics are controlled with improved substrate preparation procedures. Each of these improvements will be discussed briefly here. FILM THICKNESS Previous research identified film thickness as a critical parameter in investigating pull-off behavior [2, 11, 12, 13]. Early experimentation by Youtcheff and Aurilio to evaluate moisture sensitivity utilized glass beads of 200 µm diameter mixed with bituminous binder to control film thickness [11]. With input from UAI, Kanitpong and Bahia further modified the pull-out stubs to better control binder film thickness [3, 12]. They proposed using a smooth-surface aluminum stub and two metal support blocks to replace the glass beads. Figure 2 shows the modified stubs with metal support blocks, and Figure 3 depicts a later iteration of the pull-out stubs with aluminum frame supports.

Figure 2 – Modified pull-out stubs with metal support blocks [12].

3

nd


In early rounds of experimentation, it became clear that film thickness was still not adequately controlled with the pull-out stub support system shown in Figure 3 [14]. Figure 4 shows an improved pull-out stub designed at UWM in conjunction with UAI. Improvements to the original stub include an increase in stub diameter to 22 mm and the addition of circumferential support edges to limit the vertical position of the stub surface. Perimetrical channels in the stub edge allow excess binder to flow out from beneath the stub surface.

Figure 3 – Modified pull-out stub with aluminum frame supports [14].

Figure 4 – Modified BBS test pull-out stubs [1]. LOADING RATE Bitumen’s viscoelastic nature necessitates effective loading rate control for consistent evaluation. Research by Meng confirms that load control is critical for consistent pull-off tensile test results [2]. Early versions of PATTI were found to inadequately control loading rate so at the beginning of 2009, SEMicro launched the © PATTI Quantum Gold (PQG) test instrument that incorporated user feedback into the revised design, including improved loading rate control. The ability to control the loading rate with a graduated rate control © dial further improves consistency. The PQG comes equipped with LabView software and effectively captures load over time, allowing for calculation of the loading rate. SUBSTRATE & SURFACE ROUGHNESS Improved substrate preparation procedures involve the use of aggregate substrates that are actually used in practice. However, crushed aggregates commonly used in surface seals are not suitable substrate materials due to variations in surface roughness and texture. A procedure developed at UWM to prepare aggregate plates involves cutting large rocks into flat plates [1]. Aggregate plates are lapped with a silicon carbide compound to achieve a consistent surface texture. While aggregate plates may not fully capture aggregate surface characteristics, they represent substantial improvements over glass plates, which are still used as a control surface.

4

nd


SIGNIFICANT FACTORS INFLUENCING BOND STRENGTH DEVELOPMENT With refined substrate preparation procedures, improvements in stub design to control film thickness, and loading rate control, Miller investigated factors critical to bond strength development [1]. The factors investigated included substrate type, moisture condition, surface roughness, loading rate, curing temperature and curing humidity. An analysis of variance (ANOVA) for the screening experiment is shown in Table 1. The experiment identifies loading rate, curing temperature and humidity, and substrate type as significant main effects contributing to bond strength development. Experimental results indicate that loading rate most significantly influences the pull-out tension response. Other significant factors investigated in subsequent studies included material variables related to substrate and binder type as well as curing variables related to temperature, humidity and time. Table 1 – Analysis of variance for the factor screening experiment. Source

DF

Seq. SS

Adj. SS

Adj. MS

F

P

Substrate

1

1051.6

1051.6

1051.6

8.2

0.006

Moisture

1

0.5

0.5

0.5

0.00

0.952

Roughness

1

90.9

90.9

90.9

0.7

0.404

Loading Rate

1

31926.2

31926.2

31926.2

248.5

0.000

Temperature

1

1967.0

1967.0

1967.0

15.3

0.000

Humidity

1

1105.3

1105.3

1105.3

8.6

0.005

Error

57

7322.4

7322.4

128.5

Total

63

43463.8

S = 11.3342

R-Sq = 83.15%

R-Sq (adj) = 81.38%

LOADING RATE EXPERIMENT Loading rate is clearly identified as a factor significantly influencing the pull-out tension response. Experimental results indicated that the pull-out tension response increases as loading rate increases. A power law model adequately captures the relationship between loading rate and pull-out tension. Results also show that loading rates between 690-1030 kPa/s (100-150 psi/s) appear to exhibit a linear relationship above pull-out tension values of 690 kPa (100 psi). Loading rates exceeding 2700 kPa/s (400 psi/s) lead to increasing variability in both the pull-out tension values and loading rate. Therefore, loading rates exceeding 2700 kPa/s (400 psi/s) should be avoided to minimize experimental error in order to obtain valid results.

Figure 5 – Loading rate and pull-out tension are described by a power law model. 5

nd


CURING CONDITIONS EXPERIMENT Miller also investigated the effects of curing temperature and humidity on pull-out tension [1]. In this experiment, the loading rate was fixed at approximately 700 kPa/s for four curing conditions. Samples cured in an environmental chamber at prescribed curing intervals of 2, 6 and 24 hours. The experiment utilized a cationic rapid-setting emulsion with high viscosity (CRS-2), granite and limestone substrates, and the following curing conditions:

Samples cured at 35 °C and 30 percent relative hum idity.




An ANOVA for the curing conditions experiment, shown in Table 2, indicates that substrate type, curing temperature, and curing interval are statistically significant main effects at a 95% confidence level, while temperature-curing interval and humidity-curing interval interactive effects are also statistically significant at this confidence level. Other important results suggest that: •

Significant strength gains were observed between two and six hours at the 35 °C temperature level.

•

Granite outperformed limestone in three of four curing conditions after six hours.

•

Humidity did not significantly affect the pull-out tension response.

•

Samples tested at 35 °C and 30 percent relative hu midity performed better than samples tested at other curing conditions.

•

Samples exhibited only slight differences in the pull-out tension response after 24 hours of curing.

Table 2 – Analysis of variance for the curing conditions experiment. Source

DF

Seq. SS

Adj. SS

Adj. MS

F

P

Substrate

2

2862.9

2862.9

1431.4

16.44

0.000


1

2079.2

2079.2

2079.2

23.8 8

0.000


1

60.6

60.6

60.6

0.70

0.417

Curing Interval

2

113689.5

113689.5

56844.8

652.96

0.000


2

247.0

247.0

123.5

1.42

0.271

Substrate-Humidity

2

559.2

559.2

279.6

3.21

0.067

Substrate-Interval

4

838.9

838.9

209.7

2.41

0.092


1

54.4

54.4

54.4

0.63

0.441


2

3695.7

3695.7

1847.8

21.23

0.000

Humidity-Interval

2

1354.8

1354.8

677.4

7.78

0.004

Error

16

1392.9

1392.9

87.1

Total

35

126835.1

S = 9.33046

6

R-Sq = 98.9%

R-Sq (adj) = 97.6%

nd


EMULSION TYPE EXPERIMENT Based on results from the screening and curing conditions experiments, Miller developed a materials experiment to investigate a variety of emulsion and substrate combinations [1]. The experiment included five emulsion types and three substrate types at four curing intervals. Figure 6 and Figure 7 show results from this experiment. Two initial observations can be made in Figure 6: that both cationic rapid-setting emulsions (CRS-2 lab and CRS-2 field) perform better than polymer modified cationic rapid-setting emulsions (CRS-2P) and high-float anionic rapid setting (HFRS-2) emulsions; and that all emulsion types exhibit sharp increases in pull-out tension initially, with relative gains in tensile strength diminishing over time. A power law model adequately characterizes the relationship between curing interval and pull-out tension. In Figure 7, glass plates yield a near-perfect correlation using a power law model, with solid aggregate plates and chip substrates also yielding very strong relationships.

Figure 6 – Pull-out tension values differ for various emulsion types at a range of curing intervals [1].

Figure 7 – Pull-out tension values differ for various substrate types at a range of curing intervals [1].

7

nd


CORRELATIONS OF BBS TEST RESULTS In order to validate the BBS test procedure, Miller correlated BBS test results to two other common test procedures. To gain insight into chip retention, Miller correlated BBS test results to results obtained using the ASTM D7000 sweep test procedure [1, 15]. In the sweep test, bitumen is applied at a fixed application rate to a felt disk before aggregate chips are applied and compacted. The test apparatus brushes the sample for 1 minute in an attempt to simulate the mechanical brooming action experienced by newly-constructed chip seals. The response variable considered in the sweep test is percent aggregate loss. SWEEP TEST Establishing a correlation between BBS test results and sweep test results entailed comparing the pull-out tensile strength values obtained using the BBS test to aggregate loss measured using the sweep test at identical curing conditions. Given the existing recommended sweep test performance limit of 10 percent aggregate loss, Miller devised a test correlation between BBS and sweep test results. Sweep test samples prepared with granite and limestone aggregates and CRS-2 and CRS-2P emulsions are compared to BBS test results for similar material combinations, as shown in Figure 8. At 2 hours curing, neither the sweep test samples nor the BBS test samples exhibit good performance in terms of aggregate retention or pull-out tension as indicated by high aggregate loss in the sweep test and low pull-out tension values in the BBS test. At 6 hours curing, cohesion and adhesion become more evident with improved aggregate retention and pull-out tension values. After 24 hours curing, samples exhibit greater performance in both tests. As with other experiments, a power law model appears to adequately 2 characterize the relationship between BBS results and sweep test results, with R > 0.995. Based on this relationship, a minimum BBS specification limit of 850 kPa (123 psi) may be suggested, a limit that signifies when the binder has gained sufficient bond strength to achieve less than 10 percent aggregate loss as measured by the sweep test.

Figure 8 – When sweep test results are compared to BBS test results, a potential BBS specification limit may be proposed at 850 kPa (123 psi) to define a specification target range [1]. 8

nd


DYNAMIC SHEAR RHEOMETER (DSR) STRAIN SWEEP Strain sweep procedures developed for the dynamic shear rheometer (DSR) were also compared to BBS results [13, 16]. DSR strain sweep results were considered for two evaluation criteria. Test results were analyzed in the linear range (G* / sin δ at 12 percent strain) and non-linear range (G* / sin δ at 40 percent strain). Emulsion residues cured on granite and limestone substrates demonstrated the effect of curing temperature on strength gain, particularly at early curing intervals. None of the samples attained the full strength of the neat base binder, indicating the presence of water in the emulsion even after 24 hours of curing. BBS test results are correlated with emulsion residue properties in Figure 9. Results demonstrate a strong correlation between pull-out tensile strength and resistance to deformation (G* / sin δ) of the emulsion residue. The effect of curing time is also seen in comparing the results, with binder stiffness and bond strength increasing with time. DSR strain sweep results suggest that the BBS test is capturing a fundamental engineering property in bond strength. The BBS-DSR correlation reinforces the notion that bond strength development can be quantified as the emulsion breaks and begins to release water.

Figure 9 - BBS test results compare well to DSR strain sweep results at two strain levels.

INTER-LABORATORY EVALUATION OF BBS TEST METHOD To further validate BBS test methods with inter-lab experimentation, research personnel at the University of Stellenbosch – South Africa (USSA) conducted a series of tests based on the draft test method developed by Miller and UWM research personnel. The South African evaluation utilized an identical experimental setup (e.g. PQG testing instrument and modified pull-out stubs) for a different set of emulsified binders and substrate types. Substrate preparation procedures differed slightly due to differences in available substrate preparation equipment. UWM tests utilized large aggregate plates that accommodated four pull-out stubs, while USSA tests utilized aggregate slices from cored rock samples that accommodated only a single pullout stub. Researchers at USSA conducted initial material evaluation tests at 400 kPa/s and 900 kPa/s but later fixed the loading rate at approximately 700 kPa/s based on UWM results and recommendations. Inter-lab evaluation considered four emulsion types and two substrate types (e.g. granite and tillite). Availability of an environmental chamber at the University of Stellenbosch limited the curing conditions to 30 °C and amb ient levels of relative humidity. The following emulsions were considered for both substrate types at 2, 6 and 24 hour curing intervals: 9

nd


•

Cationic rapid setting (CRS65) bitumen emulsion.

•

Anionic slow setting (SS60) bitumen emulsion.

•

Cationic rapid setting (CRS65) bitumen emulsion modified with 3% latex.

•

Anionic slow setting (SS60) bitumen emulsion modified with 3% latex.

Figure 10 and Figure 11 depict results for different emulsion types and substrate types, respectively. These findings corroborate the findings of Miller and research personnel at UWM that 1) bond strength development can be quantified over time; 2) the BBS test method can characterize bond strength development for different emulsion types; and 3) that advantageous combinations of aggregate-binder can be identified using the BBS test method.

Figure 10 – Pull-out tension values differ for various emulsion types at a range of curing intervals.

Figure 11 – Pull-out tension values differ for various substrate types at a range of curing intervals.

10

nd


Table 3 displays experimental results from inter-laboratory testing. Values of the coefficient of variation (COV) are typically less than 10 percent, indicating good repeatability for three replicates in spite of a different substrate preparation procedure. Experimental results are also depicted in Figure 12 for unmodified bitumen and Figure 13 for modified bitumen. Table 3 –BBS test results for materials tested at the University of Stellenbosch. Experimental Factors Binder

Pull-Out Tensile Strength (kPa)

Substrate

Cure Interval (hr) 2

Granite

6

533.0

19.5

11.3

3.7%

24

919.7

46.7

26.9

5.1%

2

406.6

15.9

9.2

3.9%

6

765.2

77.2

44.6

10.1%

24

835.6

57.1

33.0

6.8%

2

482.4

13.4

7.7

2.8%

6

466.2

5.6

3.3

1.2%

24

545.7

106.9

61.7

19.6%

2

445.4

8.3

4.8

1.9%

6

459.9

9.8

5.6

2.1%

24

568.3

51.5

29.7

9.1%

2

439.1

4.1

2.4

0.9%

6

824.8

13.9

8.0

1.7%

24

989.2

169.9

98.1

17.2%

2

446.3

12.4

7.2

2.8%

6

846.5

60.4

34.9

7.1%

24

1095.8

104.8

60.5

9.6%

2

472.5

6.8

3.9

1.4%

6

505.9

12.4

7.2

2.5%

24

477.9

5.6

3.3

1.2%

2

449.9

6.8

3.9

1.5%

6

484.2

2.7

1.6

0.6%

24

717.3

52.8

30.5

7.4%

CRS65 Tillite


Granite CRS65 + 3% LATEX Tillite

Granite SS60 + 3% LATEX Tillite

Average St Dev St Error 424.6 4.7 2.7

COV 1.1%

Figure 12 shows a combination of test results for CRS65 and SS60 bitumen emulsions on granite and tillite aggregates for different loading rates. The following points of interest are noted:

11

•

The CRS65-granite combination outperforms the CRS65-tillite combination at all curing intervals. The pull-out tension appears to increase linearly with curing time.

•

SS60 results show very little strength increase over 24 hours, except on tillite at 24 hours. The result is expected from a slow setting emulsion applied on acidic aggregates such as granite and tillite. These aggregates contain silica and have a strong negative charge in the presence of water. This negative charge attracts positively charged cationic bitumen particles, leading to destabilization of the surfactant system and subsequent coagulation of the bitumen particles. This breaking mechanism is absent when anionic emulsions are used with acidic aggregates [17].

nd


•

Tests conducted at 900 kPa/s show slight increases in pull–out tensions values over those tested at 700 kPa/s, thereby confirming that higher loading rates result in higher pull-out tension values. SS60 & CRS65 on GRANITE & TILLITE 1200 SS60 @ 900 kPa/s SS60 @ 700kPa/s

1000

CRS65 @ 900kPa/s


CRS65 @ 700kPa/s

800

600

400

200


2 HOURS on GRANITE

24 HOURS on GRANITE

2 HOURS on TILLITE

6 HOURS on TILLITE

24 HOURS on TILLITE

CURING TIME

Figure 12 – BBS testing of unmodified bitumen shows bond strength development over time. SS60+3% & CRS65+3% LATEX on GRANITE & TILLITE 1200 SS60+3% @ 800 kPa/s SS60+3% @ 700kPa/s CRS65+3% @ 400 kPa/s


1000

CRS65+3% @ 950kPa/s

800

600

400

200


6 HOURS on GRANITE

24 HOURS on GRANITE

2 HOURS on TILLITE

6 HOURS on TILLITE

24 HOURS on TILLITE

CURING TIME

Figure 13 – BBS testing of modified bitumen also shows bond strength development over time.

Figure 13 shows a combination of test results for CRS65 + 3% latex and SS60 + 3% latex on granite and tillite aggregate substrates. The following points of interest are noted:

12

nd


•


•


•

After 6 hours the results for CRS65 + 3% latex-granite show higher tensile strength values than SS60 + 3% latex-granite.

•

After 6 hours, there are limited differences in the results of the CRS65 + 3% latex-tillite and SS60 + 3% latex- tillite.

•

After 24 hours, the CRS65-granite combination shows significantly higher strength values than the SS60-granite combination but smaller values that the CRS65 + 3% latex-tillite and SS60+ 3 % latextillite combinations.

•

The modified emulsions performs better that the unmodified emulsions after 24 hours.

•

The CRS65 + 3% latex-tillite and SS60 + 3% latex-tillite combinations values relate well and seem to be superior to granite.

CONCLUSION This paper reviewed the development of a new test called the Bitumen Bond Strength (BBS) test and its application to bituminous emulsions. Inter-laboratory test results reinforce the hypothesis that the BBS test protocol can be used to effectively evaluate bond strength of different emulsion types and substrate types. Aside from loading rate, emulsion type and curing interval are identified as the most significant factors contributing to bond strength development. Substrate type is also identified as a significant factor leading to bond strength. Interactions between emulsion type and curing interval are identified as the most significant interaction. Further validation of the BBS test method is needed for the test to be integrated into a performance-based specification system for surface seals.

REFERENCES [1]

Miller, T. “Development of Bond Strength Test for Improved Characterization of Asphalt Emulsions.” Master of Science thesis, University of Wisconsin – Madison, 2010.

[2]

Meng, J. “Affinity of Asphalt to Mineral Aggregate: Pull-Off Test Evaluation.” Master of Science thesis, University of Wisconsin – Madison, 2010.

[3]

Canestrari, F., Cardone, F., Graziani, A., Santagata, F., and Bahia, H. “Adhesive and Cohesive Properties of Asphalt-Aggregate Systems Subjected to Moisture Damage.” Proc., European Asphalt Technology Association, Journal of Road Materials and Pavement Design, 11-32, 2010.

[4]

Bahia, H., Meng, J., Velasquez, R., Miller, T., Daranga, C., and Moraes, R. “Evaluation of the Bitumen Bond Strength (BBS) Test for Moisture Damage Characterization.” Asphalt Research Consortium report prepared for the Federal Highway Administration, University of Wisconsin – Madison, 2010.

[5]

Opus International Consultants. “Performance Based Bitumen Specification.” Discussion paper, New Zealand, 2010.

[6]

Epps, A., Glover, C., and Barcena, R. “A Performance-Graded Binder Specification for Surface Treatments.” Report FHWA/TX-02/1710-1, Texas Department of Transportation, 2001.

[7]

Comité Européen de Normalisation (CEN). “Activity Report of the CEN Ad-Hoc Group Adhesion/Durability.” Report N090, 2009.

[8]

James, A. “Overview of Asphalt Emulsions.” Circular E-C102, Transportation Research Board, Washington, D.C., 2006.

[9]

Redelius, P. and Walter, J. “Emulsion and Emulsion Stability.” Taylor & Francis, New York, 2006.

13

nd


[10]

ASTM. “ASTM D4541-09 Standard Test Method for Pull-Off Strength of Coatings Using Portable Adhesion Testers.” American Society of Testing and Materials, West Conshohocken, PA, 2009.

[11]

Youtcheff, J., Williams, C. and Stuart, K. “Moisture Sensitivity Testing of Bitumen Using a Pneumatic Adhesion Test.” Eurobitume Workshop, Luxembourg, 1999.

[12]

Kanitpong, K., Bahia, H. “Role of Adhesion and Thin Film Tackiness of Asphalt Binders in Moisture Damage of HMA.” J. Asphalt Paving Technology, 72, 502-528, 2003.

[13]

Miller, T., Arega, Z., and Bahia, H. “Correlating Rheological and Bond Properties of Emulsions to Aggregate Retention of Chip Seals.” Proc., Transportation Research Record 1860, Transportation Research Board, Washington, D.C., in press, 2010.

[14]

Fratta, D. and Daranga, C. “Experimental Validation of the Modified PATTI Test Methodology.” Asphalt Research Consortium report prepared for the Federal Highway Administration, University of Wisconsin – Madison, 2006.

[15]

ASTM. ASTM D7000-08 Standard test method for sweep test of bituminous emulsion surface treatment samples. American Society for Testing and Materials, West Conshohocken, PA, 2004.

[16]

Arega, Z. “Rheological Characterization of Asphalt Emulsions for Chip Seal Applications.” Master of Science thesis, University of Wisconsin – Madison, 2009.

[17]

Louw, K., Spence, K. and Kuun, P. “The use of bitumen emulsions as a cost effective solution for constructing seals during winter.” Proc., 8th Conference on Asphalt Pavements for Southern Africa (CAPSA), Sun City, South Africa, 2004.

AUTHOR BIOGRAPHIES Timothy Miller is a research assistant and teaching assistant in the Department of Civil and Environmental Engineering at the University of Wisconsin – Madison. Mr. Miller is pursuing his Ph.D. degree at UWMadison and leading efforts to improve frameworks for analyzing asphalt pavement sustainability. André Greyling is a professionally registered civil engineer working in the consulting engineering field in Cape Town, South Africa. He is currently completing a part-time Msc. Eng degree in pavement engineering under the guidance of Professor Kim Jenkins at the University of Stellenbosch. Dr. Hussain Bahia is a professor of Civil and Environmental Engineering at the University of WisconsinMadison. Prior to joining the University of Wisconsin in 1996, he served as a research assistant, research associate, and Assistant Professor at Pennsylvania State University. Dr. Bahia has served as a principal investigator and co-investigator for a number of research projects and is the author/co-author of many research papers related to asphalt binders, mixtures, and pavement construction. Dr. Kim Jenkins studied Civil Engineering at the University of Natal in South Africa and worked for 12 years in the consulting engineering environments. He focused on geotechnics, road materials and pavement design, specializing in road rehabilitation and recycling technology. Since 1996 he has worked at Stellenbosch University and is currently the incumbent SANRAL Chair in Pavement Engineering. Copyright License Agreement The Author allows ARRB Group Ltd to publish the work/s submitted for the 2 Conference, granting ARRB the non-exclusive right to:

nd

International Sprayed Sealing

• publish the work in printed format • publish the work in electronic format • publish the work online. The author retains the right to use their work, illustrations (line art, photographs, figures, plates) and research data in their own future works The Author warrants that they are entitled to deal with the Intellectual Property Rights in the works submitted, including clearing all third party intellectual property rights and obtaining formal permission from their respective institutions or employers before submission, where necessary.

14

nd


APPENDIX: DRAFT STANDARD OF BBS TEST PROCEDURE Introduction A draft standard for the bitumen bond strength (BBS) test has been developed and submitted for review to the American Association of State Highway and Transportation Officials (AASHTO). This section will describe and outline the draft procedure in its current form.

Scope of the Proposed Method This test method quantifies the tensile force needed to remove a pull-out stub from an asphalt binder adhered to a solid substrate. Samples are conditioned at a controlled temperature and humidity based on experimental conditions determined at the beginning of the investigation. A pneumatic load is applied to the pullout stub until failure using an ASTM Type IV adhesion tester in accordance with the standard ASTM D4541 method. The pullout tension at failure is used to describe the adhesive properties of asphalt binders and the compatibility between particular aggregates and binders. This test may involve hazardous materials, operations, and equipment. This test method does not purport to address all of the safety concerns associated with its use. It is the responsibility of the user to establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use.

Summary of Method An adhesion tester applies a pneumatic load via a pressure ring to a pull-out stub fixed to the substrate with asphalt binder, as shown in Figure 14. The binder is adhered to the substrate and can be subjected to conditioning at fixed levels of temperature and humidity prior to testing. Moisture conditioning is also possible for hot-applied binders to evaluate the effects of moisture damage. Pull-out stubs are allowed to acclimate to laboratory conditions prior to testing. The stress applied to the assembly is recorded over time, allowing for calculation of load to failure and loading rate. The surface of the substrate is visually examined to determine the type of failure mode.

Figure 14 – Schematic representation of the BBS test assembly. Scale is approximate.

15

nd


Significance and Use •

Pull-out tensile strength values measured over a range of environmental conditions and curing times provide information related to the adhesive and cohesive behavior of hot and emulsified asphalt binders.

•

Evaluation of pull-out tensile strength on different aggregate substrates can be used to evaluate asphalt-aggregate compatibility.

•

For emulsified binders, materials may be compared based on curing rate and ultimate tensile strength.

•

For hot binders, the pull-out tensile strength may be used to evaluate moisture damage as measured by the decrease in tensile strength due to moisture conditioning for given asphalt-aggregate combinations.

Apparatus Molds – For emulsified binders, use a silicone mold measuring approximately 40 mm by 40 mm with a 20 mm-diameter hole and 0.8 mm thickness. The mold has no backing and is used to contain the emulsion on the aggregate surface. For hot binders, use a silicone mold measuring approximately 40 mm by 40 mm with a 10 mm-diameter cavity. This mold is similar to the molds used to prepare 8-mm dynamic shear rheometer (DSR) test samples. Figure 15 depicts diagrams of each mold type.

Figure 15 – Mold dimensions (mm) for emulsified binders (left) and hot binders (right). Base Plate – A solid aggregate substrate, aggregate composite, or glass plate of sufficient thickness is used to maintain structural integrity during testing. For emulsified binders, the plate must be uniformly flat to ensure that the binder will not flow beneath the mold during curing. For both binder types, the plate must be uniformly flat to reduce the possibility of eccentric loading when the pullout stub is tested. Testing Machine – Use a Type IV adhesion tester as defined by ASTM D4541 for all tests. Such a device must consist of, at minimum, a control module, pressure ring, pressure plate, and data capture software. While different loading fixtures are available, the F-2 size pressure ring and pressure plate work sufficiently well for both types of asphalt binders. Figure 16 depicts an example testing machine.

16

nd


Rate Control

Test Button Reaction Plate Loading Direction

Piston Pullout Stub Piston Support Test Sample

Substrate

Figure 16 – General representation of the BBS test apparatus. Air Supply – Capable of producing a consistent air pressure of at least 0.67 MPa (100 psi) as read on the supply gauge. Self-contained air cylinders, shop (bottled) air, or air from an automatic pump may be used. Pull-out Stubs – Stainless steel or any other suitable material, with dimensions in mm as shown in Figure 17. Stub edges are beveled to reduce the amount of binder trapped between the stub edge and substrate and to ensure a uniform film thickness.

40 7

15

6

22 20

5

0.8

Top View

Bottom View

Figure 17 – Pull-out stub dimensions (mm) for the BBS test. Forced Draft Oven – Capable of maintaining temperatures of at least 150 +/- 3 °C for preparing all aggregate and binder samples. Two temperature-controlled ovens of Type IIA or IIB as defined in ASTM E145 should be used due to different heating conditions required in the sample preparation process. Environmental Chamber – Capable of maintaining temperatures between 15 °C a nd 75 °C +/- 1°C, and relative humidity between 20 percent and 80 percent +/- 1 percent, for curing all emulsion samples. Thermometer – For tests performed at 25 °C (77°F), use ASTM Therm ometer No. 17C (17F) to measure the temperature of the aggregate surface prior to testing. For tests performed at other temperatures, use ASTM thermometers of an appropriate range and accuracy equal to that of the No. 17 thermometer. Since the accuracy of the test results is dependent upon closely controlled temperature conditions, thermometers

17

nd


should be calibrated in accordance with ASTM E77. Thermometric devices with the same accuracy as ASTM thermometers may also be used. Container – Any suitable container may be used to hold the hot asphalt material while being melted. For emulsified binders, the suitable container may be plastic, non-metal, or epoxy-lined, if metal. Ultrasonic Cleaner – An ultrasonic cleaner is needed to remove residual particles from substrate surfaces prior to testing. The ultrasonic cleaner should be capable of maintaining temperatures of 60 °C and s hould have a chamber large enough to allow for complete submersion of the substrate.

Aggregate Test Specimens Solid Aggregate Substrates – Cut aggregate substrates from either quarried rocks or cored rock samples using standard rock saws such that plate faces are parallel. Lap all substrates using a 280-grit silicon carbide material on a standard lapidary wheel to remove saw marks and to ensure a consistent surface roughness. Once cut, clean samples for 60 minutes in an ultrasonic cleaner containing distilled water at a temperature of 60 °C to remove residual particles o n the plate surface. Composite Substrates – Composite substrates may be prepared by cutting slices of samples cast in standard Portland cement concrete cylinder molds. Composite substrates contain aggregate chips and a rapid-setting cement compound. These substrates need not be lapped but may be cleaned in the ultrasonic cleaner to remove dust and residual particles.

Emulsified Binders Sample Preparation:

18

•

Obtain a representative sample of the material for testing using procedures specified in ASTM D140.

•

Heat emulsified binder to an application temperature of 60 ± 2 °C in a forced-draft oven. Heat emulsion in a covered plastic container or non-metallic container for no longer than 1.5 hours to avoid premature breaking.

•

Simultaneously heat aggregate substrates to an application temperature of 25 ± 2 °C in a second forced-draft oven.

•

If moisture conditioning samples, allow the substrates to soak in a bath of heated distilled water to achieve the saturated surface dry (SSD) condition while preheating.

•

Place molds without backing on the substrate surface and fill with liquid emulsion using a plastic eyedropper. Binder samples should be positioned on the substrate to allow for placement of multiple samples and for sufficient clearance of the testing apparatus.

•

Cure the substrate and filled molds under controlled conditions in an environmental chamber for a given curing interval.

•

While samples are curing, heat pullout stubs to 60 ± 2 °C in a forced-draft oven.

•

After removing samples from the environmental chamber, remove the silicone molds encircling the cured binder and place the heated pull-out stubs on the binder samples.

•

Firmly press pull-out stub down on substrate surface to ensure good adhesion between binder, stub, and substrate. Expect excess binder to flow out of the pull-out stub channels. Avoid twisting the pull-out stub to reduce air entrapment between the binder sample and substrate surface.

•

Return the testing assembly to a forced draft oven at 25 ± 2 °C for approximately 1 hour to allow the samples to acclimate to testing conditions.

nd


Test Procedure: •


•

Place a shaft collar or four spacers on opposite sides of the sample to be tested to reduce eccentric loading and to allow the stub to fully separate from the substrate surface.

•

Gently place pressure ring around pull-out stub and resting on the shaft collar or spacers so as not to disturb the stub or induce unnecessary strain in the sample.

•

Gently screw pressure plate onto the pull-out stub, taking care not to rotate the stub, until the pressure plate just touches the gasket.

•


•


•

Record the maximum pull-out tension and observe the failure mode. If more than 50 percent of the substrate surface remains exposed following removal of the pull-out stub, the failure is considered adhesive. If less than 50 percent of the substrate surface remains exposed following the removal of the pull-out stub, the failure is considered cohesive.

•

Repeat the test procedure until all samples have been tested. A minimum of three samples should be tested at each set of experimental conditions.

Report

19

•

Obtain the pull-out tension directly from the graphical computer interface and report to the nearest 0.1 kPa. Environmental conditions (e.g. temperature) should be noted for each test.

•

Pull-out tension results should be averaged for each set of samples tested. A minimum of three samples should be tested at each set of experimental conditions.

•

Determine the loading rate by calculating the slope of the line between the initial pullout tension and the final failure tension. The slope of the line should be calculated between 20 percent and 80 percent of the maximum stress. For best results, the loading rate should be 700 – 1400 kPa/s.

•

Document the failure mode as cohesive, adhesive, or a combination of failure modes through visual observation and photo analysis. Reject samples that exhibit adhesive failure between the pullout stub surface and asphalt binder due to inadequate stub coverage.


ADDENDUM 6 – PAPER PRESENTED AT 10TH CONFERENCE ON ASPHALT PAVEMENT FOR SOUTHERN AFRICA, 2011

124 AH GREYLING

December 2012

10th CONFERENCE ON ASPHALT PAVEMENTS FOR SOUTHERN AFRICA

DEVELOPMENT OF A TEST METHOD FOR DETERMINING EMULSION BOND STRENGTH USING THE BITUMEN BOND STRENGTH (BBS) TEST - A SOUTH AFRICAN PERSPECTIVE Andre Greyling, University of Stellenbosch, South Africa Timothy Miller, University Wisconsin –Madison, United States of America Prof. Hussain Bahia, University of Wisconsin – Madison, United States of America Prof. Kim Jenkins, University of Stellenbosch, South Africa

Abstract The need to understand and scientifically quantify the expected behavior of surfacing seals has been around for as long as bitumen and aggregate have been used to surface roads. The design of surfacing seals has generally been based on available materials, skills and current client preferences. Tests performed on surfacing seal aggregate and bitumen binders are normally limited to ensure that the separate materials comply with their relative classification specifications. During the seal design process, no tests are conducted to quantify the bond strength that develops between various binders and aggregate combinations. Developing practical and effective test methods to characterize this bond strength is therefore critical for pavement designers and contractors to predict how a surface seal will perform over the expected service life. Research conducted at the University of Wisconsin – Madison (UWM) identified the Pneumatic Adhesive Tensile Testing Instrument, commonly known as PATTI, as an appropriate instrument for evaluating bond strength development in the newly developed Bitumen Bond Strength (BBS) test. (AASHTO TP-91, 2011) Recently developed at the UWM in conjunction with the University of Ancona – Italy and the University of Stellenbosch – South Africa, the BBS test procedure has been used successfully to characterize moisture sensitivity and bond strength development in hotapplied binders and bitumen emulsions, respectively (Miller 2010). This paper describes the test method and evaluates the relevance of the procedure using bitumen and bitumen emulsions produced in South Africa and U.S.A. 1- INTRODUCTION Chip and spray surfacing seal design has experienced renewed interest and continuous improvement and development is several countries over the past two decades. In South Africa, seals are continually used as more and more attention is given to the periodic maintenance of existing surfaced roads. There is also a significant increase in the use of surfacing seals in North America as the need to develop more energy and resource efficient surfacing options becomes a priority. Despite this growing surface seal use, the selection of binder type and grade does not always follow rigorous scientific processes. Most seals are designed based on experience and relying on common practice and best judgment. Historically, empirical information and contractor experience strongly influence surface seal design.

1


With increasing traffic volumes and heightened performance demands, performance-based specifications are being developed in the USA and Europe to account for deficiencies in surface seal design methodologies (Bahia et al, 2010;Opus International Consultants, 2010) Performancebased specifications for surface seals identify the need for a simple and inexpensive technique for evaluating emulsion bond strength development over time as well as binder-aggregate compatibility. Although various tests exist for investigating adhesion between bituminous emulsions and aggregate chips (Comité Européen de Normalisation, 2009), some of these tests have limited application to performance-based specifications. The Bitumen Bond Strength (BBS) test method aims to address some of the limitations encountered in evaluating bond strength. Researchers at the University of Wisconsin – Madison (UWM) recently developed the BBS test (AASHTO TP-91, 2011) in partnership with the University of Ancona – Italy (UAI) and the University of Stellenbosch – South Africa (USSA) for hot applied binders and emulsions, respectively. USSA became involved in BBS test efforts in 2008 to assist in the development and evaluation of the BBS test method, while UAI contributed significantly to the development of the test apparatus over the past four years. Due to limited time and resources, the involvement of the USSA was limited to various discussion sessions, the evaluation of the BBS test, and conducting a series of control tests. The developers of the BBS test envision that the test method will be able to quantify and characterize bond strength development, adhesive properties and aggregate-binder compatibility, thereby improving the effectiveness of surface seal design. Further research is needed to identify other parameters, such as traffic-related loading rates, in order to more accurately simulate field distresses encountered in early-life phases of surface seals. Bond strength development is critical for new surface seals as stone loss and loose aggregate must be minimized once a road enters into service following construction. Bond strength development depends on several factors, including the material characteristics of the emulsion, environmental conditions during construction, and the mineralogical characteristics of the aggregate chips (James, 2006; Redelius, and Walter, 2006). The rate at which bond strength develops directly affects an acceptable time threshold at which the seal may be opened to traffic. The seal must have sufficient bond strength to retain aggregate chips, but the seal must also gain this strength in a sufficiently short period of time to minimize user delay. Current evaluation and validation practices for certifying newly applied chip seals often rely on subjective judgment, though various specification systems are under development. Because bond strength development between emulsions and aggregate chips in early-life phases has implications for early raveling, developing a standard test method that evaluates bond strength is critical for reducing premature seal failures. This paper is a revision of a paper published by Miller et al in 2010 at the 2nd International Spray Sealing Conference and focuses on the development of the test procedure and the results of tests conducted at USSA and UWM. The test was specifically conducted to characterize bitumen emulsions commonly used in surfacing seals.

2


2- BBS TEST DEVELOPMENT 2.1

Background

The need to understand and scientifically quantify the expected behavior of surfacing seals has been around for as long as bitumen and aggregate have been used to surface roads. The design of surfacing seals has generally been based on available materials, skills and current client preferences. Tests performed on surfacing seal aggregate and bitumen binders are generally limited to ensuring that the separate materials comply with their relative classification specifications. During the seal design process, no tests are conducted to quantify the bond strength that develops between binders and aggregate combinations. Developing practical and effective test methods to characterize this bond strength is therefore critical for pavement designers and contractors to predict how a surface seal will perform over the expected service life. Pull-off tests have been widely used and developed in other industries for various test procedures. It was found that these tests could be adapted for evaluation of adhesion in surfacing seal materials. The painting industry initially developed pull-off tests to evaluate the pull-off strength of coatings on rigid substrates such as metal, concrete and wood. The goal of such pull-off test methods is to measure the maximum normal force that a solid surface coating can withstand before the adhesive detaches from the surface at failure. Such tests allow for the evaluation of the failure type (e.g. adhesive or cohesive) through inspection of the failure surface after detachment has occurred (Meng, 2010; ASTM D4541-, 2009). 2.2

ORIGINAL PATTI TEST

The bitumen industry first utilized the Pneumatic Adhesive Tensile Testing Instrument, or PATTI, in the late 1990s to evaluate adhesive loss of binder-aggregate systems exposed to moisture conditioning. (Kanitpong & Bahia, 2003) Recent research at UWM continues to investigate the effects of moisture damage for hot applied binders. (Meng, 2010; Bahia and Meng, 2010; Kanitpong & Bahia, 2003) Initial tests identified several significant effects, including variations in preparing the test assembly (operator dependence), binder film thickness, and curing and testing temperatures. Recent generations of the PATTI, notably the PATTI Quantum Gold (PQG), address some of these these shortcomings while ensuring compliance with surface seal industry requirements (Miller, 2010). Further modifications and the development of a new BBS Testing procedure (AASHTO TP-91, 2011) made the testing of the bitumen aggregate bond a reality. Early generations of the PATTI consisted of a pressure hose, adhesion tester, piston, reaction plate and a metal pull-out stub. Figure 1 below shows a typical assembly.

3


Figure 1 – Original PATTI assembly. The original test procedure entailed the following steps: • • • •

•

A hot bitumen sample is applied to a glass substrate and allowed to cure for a fixed time interval. A metal pull-out stub is applied to the bitumen sample and allowed to set for a given time interval. After placing the piston over the pull-out stub, the reaction plate is fixed to the stub. The pressure hose introduces compressed air to the piston, resulting in an upward force on the specimen and eventual failure of the binder. Failure occurs when the applied pressure exceeds the cohesive strength of the binder or the adhesive strength of the binderaggregate interface. The pressure at failure is recorded and the procedure is repeated for other test specimens.

While the original test procedure did yield quantitative information related to bond strength characteristics and failure behavior, research at UWM and UAI identified several factors influencing the effectiveness of the test method. Some of these factors include: • •

• • •

The binder film thickness between the stub and substrate could not be controlled easily. Because the original pull-out stub measured only 12.7 mm in diameter, the stub geometry limited the measurement of smaller tensile strengths. Recent modifications to the pull-out stub design at UWM and UAI improved the geometry by nearly doubling the stub diameter. The device did not report pressure over time, making the calculation of loading rate difficult. Rather, PATTI reported only the real-time applied load but not within a computerbased graphical user interface. The loading rate varied and could not be set easily. While the PATTI is equipped with a rate control dial, the dial did not effectively control the loading rate or report the real-time loading rate. Initial tests were performed on glass substrates, hardly a suitable surface seal material.

4


2.3

INITIAL TEST METHOD LIMITATIONS & MODIFICATIONS

The original PATTI test was developed for non-viscous paint and had limitations which were addressed in various ways. Improvements to the pull-out stub design, loading rate control and substrate preparation procedures represented significant advancements from the original PATTI test to its current form as the BBS test method. Binder film thickness is effectively controlled in the BBS test with an improved stub design. Loading rate is effectively controlled with new functions of the PATTI Quantum Gold. Substrate surface characteristics are controlled with improved substrate preparation procedures. Each of these improvements will be discussed briefly below. 2.3.1 FILM THICKNESS Previous research identified film thickness as a critical parameter in investigating pull-off behavior. (Meng, 2010; Youtcheff et al, 1999; Kanitpong & Bahia, 2003; Miller et al, 2010). Early experimentation by Youtcheff and Aurilio to evaluate moisture sensitivity utilized glass beads of 200 μm diameter mixed with bituminous binder to control film thickness (Youtcheff et al, 1999). With input from UAI, Kanitpong and Bahia further modified the pull-out stubs to better control binder film thickness (Canestrari, 2010; Kanitpong & Bahia, 2003). They proposed using a smoothsurface aluminum stub and two metal support blocks to replace the glass beads. Figure 2 shows the modified stubs with metal support blocks, and Figure 3 depicts a later iteration of the pull-out stubs with aluminum frame supports.

Figure 2 – Modified pull-out stubs with metal support blocks (Kanitpong & Bahia, 2003). In early rounds of experimentation, it became clear that film thickness was still not adequately controlled with the pull-out stub support system shown in Figure 3 (Fratta & Daranga, 2006). Figure 4 shows an improved pull-out stub designed at UWM in conjunction with UAI. Improvements to the original stub include an increase in stub diameter to 22 mm and the addition of circumferential support edges to limit the vertical position of the stub surface. Perimetrical channels in the stub edge allow excess binder to flow out from beneath the stub surface.

5


Figure 3 – Modified pull-out stub with aluminum frame supports.

Figure 4 – Modified BBS test pull-out stubs . 2.3.2 LOADING RATE Bitumen’s viscoelastic nature necessitates effective loading rate control for the consistent evaluation of the pull-off tensile strength. Research by Meng confirms that load control is critical for consistent pull-off tensile test results (Meng, 2010).Early versions of PATTI were found to inadequately control loading rate so at the beginning of 2009, SEMicro launched the PATTI Quantum Gold© (PQG) test instrument that incorporated user feedback into the revised design, including improved loading rate control. The ability to control the loading rate with a graduated rate control dial further improves consistency. The PQG comes equipped with LabView© software and effectively captures load over time, allowing for calculation of the loading rate. 2.3.3 SUBSTRATE & SURFACE ROUGHNESS Improved substrate preparation procedures involve the use of aggregate substrates that are actually used in practice. However, crushed aggregates commonly used in surface seals are not suitable substrate materials due to variations in in shape, surface roughness and texture. A procedure developed at UWM to prepare aggregate plates involves cutting large rocks into flat plates (Miller, 2010). The aggregate plates and disks are then lapped with a silicon carbide compound to achieve a consistent surface texture. While aggregate plates and disks may not fully capture aggregate surface characteristics, they represent substantial improvements over glass plates, which are still used as a control surface. The smooth aggregate surfaces are seen and treated as a worst case scenario.

6


2.4

SIGNIFICANT FACTORS INFLUENCING BOND STRENGTH DEVELOPMENT

With the refined substrate preparation procedures, improvements in stub design to control film thickness, and loading rate control, Miller investigated factors critical to bond strength development (Miller, 2010) The factors investigated included substrate type, moisture condition, surface roughness, loading rate, curing temperature and curing humidity. An analysis of variance (ANOVA) for the screening experiment is shown in Table 1. The experiment identifies loading rate, curing temperature and humidity, and substrate type as significant main effects contributing to bond strength development. Experimental results indicate that loading rate most significantly influences the pull-out tension response. Other significant factors investigated in subsequent studies included material variables related to substrate and binder type as well as curing variables related to temperature, humidity and time. Table 1 – Analysis of variance for the factor screening experiment. Source

DF

Seq. SS

Adj. SS

Adj. MS

F

P

Substrate

1

1051.6

1051.6

1051.6

8.2

0.006

Moisture

1

0.5

0.5

0.5

0.00

0.952

Roughness

1

90.9

90.9

90.9

0.7

0.404

Loading Rate

1

31926.2

31926.2

31926.2

248.5

0.000

Temperature

1

1967.0

1967.0

1967.0

15.3

0.000

Humidity

1

1105.3

1105.3

1105.3

8.6

0.005

Error

57

7322.4

7322.4

128.5

Total

63

43463.8

S = 11.3342

R-Sq = 83.15%

R-Sq (adj) = 81.38%

2.4.1 LOADING RATE EXPERIMENT Loading rate was clearly identified as a factor significantly influencing the pull-out tension response. Miller did extensive testing and proved that a power law model adequately captures the relationship between loading rate and pull-out tension. The results in Figure 5 show that loading rates between 690-1030 kPa/s (100-150 psi/s) appear to exhibit a linear relationship above pullout tension values of 690 kPa (100 psi). Loading rates exceeding 2700 kPa/s (400 psi/s) lead to increasing variability in both the pull-out tension values and loading rate. Therefore, loading rates exceeding 2700 kPa/s (400 psi/s) should be avoided to minimize experimental error in order to obtain valid results. Initial test done by the USSA confirmed these results. Figure 6 shows the influence of loading rates on two different aggregates and emulsion combinations. The figure clearly indicates that there is an increase in the pull out tension as the rate of application increases. The results are however limited to the 400 kPa/s to 950 kPa/s range.

7


Figure 6 – Loading rate and pull-out tension are described by a power law model.

Figure 5 – Loading rate and pull-out tension results USSA

8


2.4.2 CURING CONDITIONS EXPERIMENT The effect of curing temperature and humidity plays a very important role in the bond strength development between bitumen and aggregate. Miller investigated the effects of curing temperature and humidity on pull-out tension (Miller, 2010) In his experiment, the loading rate was fixed at approximately 700 kPa/s for four curing conditions. Samples cured in an environmental chamber at prescribed curing intervals of 2, 6 and 24 hours. The experiment utilized a cationic rapid-setting emulsion with high viscosity (CRS-2), granite and limestone substrates, and the following curing conditions: Samples cured at 35 °C and 30 percent relative humidity. Samples cured at 35 °C and 70 percent relative humidity. Samples cured at 15 °C and 30 percent relative humidity. Samples cured at 15 °C and 70 percent relative humidity. An ANOVA for the curing conditions experiment, shown in Table 2, indicates that substrate type, curing temperature, and curing interval are statistically significant main effects at a 95% confidence level, while temperature-curing interval and humidity-curing interval interactive effects are also statistically significant at this confidence level. Other important results suggest that: • • • • •

Significant strength gains were observed between two and six hours at the 35 °C temperature level. Granite outperformed limestone in three of four curing conditions after six hours. Humidity did not significantly affect the pull-out tension response. Samples tested at 35 °C and 30 percent relative humidity performed better than samples tested at other curing conditions. Samples exhibited only slight differences in the pull-out tension response after 24 hours of curing. Table 2 – Analysis of variance for the curing conditions experiment. Source

DF

Seq. SS

Adj. SS

Adj. MS

F

P

Substrate

2

2862.9

2862.9

1431.4

16.44

0.000


1

2079.2

2079.2

2079.2

23.88

0.000


1

60.6

60.6

60.6

0.70

0.417

Curing Interval

2

113689.5

113689.5

56844.8

652.96

0.000


2

247.0

247.0

123.5

1.42

0.271

Substrate-Humidity

2

559.2

559.2

279.6

3.21

0.067

Substrate-Interval

4

838.9

838.9

209.7

2.41

0.092


1

54.4

54.4

54.4

0.63

0.441


2

3695.7

3695.7

1847.8

21.23

0.000

Humidity-Interval

2

1354.8

1354.8

677.4

7.78

0.004

Error

16

1392.9

1392.9

87.1

Total

35

126835.1

S = 9.33046

R-Sq = 98.9%

9

R-Sq (adj) = 97.6%


2.4.3 EMULSION TYPE EXPERIMENT Miller developed a detailed materials experiment to investigate a variety of emulsion and substrate combinations (Miller, 2010) The experiment included five emulsion types and three substrate types at four curing intervals. Figure 8 and Figure 9 show results from this experiment. Two initial observations can be made in Figure 6: that both cationic rapid-setting emulsions (CRS-2 lab and CRS-2 field) perform better than polymer modified cationic rapid-setting emulsions (CRS2P) and high-float anionic rapid setting (HFRS-2) emulsions; and that all emulsion types exhibit sharp increases in pull-out tension initially, with relative gains in tensile strength diminishing over time. A power law model adequately characterizes the relationship between curing interval and pull-out tension. In Figure 8, glass plates yield a near-perfect correlation using a power law model, with solid aggregate plates and chip substrates also yielding very strong relationships.

Figure 7 – Pull-out tension values differ for various emulsion types at a range of curing intervals (Miller, 2010)

Figure 8 – Pull-out tension values differ for various substrate types at a range of curing intervals (Miller, 2010)

10


The USSA conducted test on granite and tillite aggregates using various modified and unmodified emulsion combinations. The results of these tests correlate well with the results from UWM and are shown in the Figure 9 below:

Figure 9 – Pull-out tension values differ for various emulsion types at a range of curing intervals

2.4.4 CORRELATIONS OF BBS TEST RESULTS In order to validate the BBS test procedure, Miller correlated BBS test results to two other common test procedures. To gain insight into chip retention, Miller correlated BBS test results to results obtained using the ASTM D7000 sweep test procedure (Miller, 2010; ASTM D7000, 2008). In the sweep test, bitumen is applied at a fixed application rate to a felt disk before aggregate chips are applied and compacted. The test apparatus brushes the sample for 1 minute in an attempt to simulate the mechanical brooming action experienced by newly-constructed chip seals. The response variable considered in the sweep test is percent aggregate loss. SWEEP TEST Establishing a correlation between BBS test results and sweep test results entailed comparing the pull-out tensile strength values obtained using the BBS test to aggregate loss measured using the sweep test at identical curing conditions. Given the existing recommended sweep test performance limit of 10 percent aggregate loss, Miller devised a test correlation between BBS and sweep test results. Sweep test samples prepared with granite and limestone aggregates and CRS-2 and CRS-2P emulsions are compared to BBS test results for similar material combinations, as shown in Figure 10.

11


At 2 hours curing, neither the sweep test samples nor the BBS test samples exhibit good performance in terms of aggregate retention or pull-out tension as indicated by high aggregate loss in the sweep test and low pull-out tension values in the BBS test. At 6 hours curing, cohesion and adhesion become more evident with improved aggregate retention and pull-out tension values. After 24 hours curing, samples exhibit greater performance in both tests. As with other experiments, a power law model appears to adequately characterize the relationship between BBS results and sweep test results, with R2 > 0.995. Based on this relationship, a minimum BBS specification limit of 850 kPa (123 psi) may be suggested, a limit that signifies when the binder has gained sufficient bond strength to achieve less than 10 percent aggregate loss as measured by the sweep test.

Figure 10 – When sweep test results are compared to BBS test results, a potential BBS specification limit may be proposed at 850 kPa (123 psi) to define a specification target range (Miller,2010) DYNAMIC SHEAR RHEOMETER (DSR) STRAIN SWEEP Strain sweep procedures developed for the dynamic shear rheometer (DSR) were also compared to BBS results (Miller et al, 2010; Arega, 2009) . DSR strain sweep results were considered for two evaluation criteria. Test results were analyzed in the linear range (G* / sin δ at 12 percent strain) and non-linear range (G* / sin δ at 40 percent strain). Emulsion residues cured on granite and limestone substrates demonstrated the effect of curing temperature on strength gain, particularly at early curing intervals. None of the samples attained the full strength of the neat base binder, indicating the presence of water in the emulsion even after 24 hours of curing.

12


BBS test results are correlated with emulsion residue properties in Figure 11. Results demonstrate a strong correlation between pull-out tensile strength and resistance to deformation (G* / sin δ) of the emulsion residue. The effect of curing time is also seen in comparing the results, with binder stiffness and bond strength increasing with time. DSR strain sweep results suggest that the BBS test is capturing a fundamental engineering property in bond strength. The BBS-DSR correlation reinforces the notion that bond strength development can be quantified as the emulsion breaks and begins to release water.

Figure 11 - BBS test results compare well to DSR strain sweep results at two strain levels.

3- INTER-LABORATORY EVALUATION OF BBS TEST METHOD To validate BBS test methods with inter-lab experimentation, research personnel at the University of Stellenbosch – South Africa (USSA) conducted a series of tests based on the draft test method developed by Miller and UWM research personnel. The South African evaluation utilized an identical experimental setup (e.g. PQG testing instrument and modified pull-out stubs) for a different set of emulsified binders and substrate types. Substrate preparation procedures differed slightly due to differences in available substrate preparation equipment. UWM tests utilized large aggregate plates that accommodated four pullout stubs, while USSA tests utilized aggregate slices from cored rock samples that accommodated only a single pull-out stub. Researchers at USSA conducted initial material evaluation tests at 400 kPa/s and 900 kPa/s but later fixed the loading rate at approximately 700 kPa/s based on UWM results and recommendations. Inter-lab evaluation considered four emulsion types and two substrate types (e.g. granite and tillite). Availability of an environmental chamber at the University of Stellenbosch limited the curing conditions to 30 °C and ambient levels of relative humidity. The following emulsions were considered for both substrate types at 2, 6 and 24 hour curing intervals: • Cationic rapid setting (CRS65) bitumen emulsion. • Anionic slow setting (SS60) bitumen emulsion. • Cationic rapid setting (CRS65) bitumen emulsion modified with 3% latex. • Anionic slow setting (SS60) bitumen emulsion modified with 3% latex.

13


Figure 12 and Figure 13 depict results for different emulsion types and substrate types, respectively. These findings corroborate the findings of Miller and research personnel at UWM that 1) bond strength development can be quantified over time; 2) the BBS test method can characterize bond strength development for different emulsion types; and 3) that advantageous combinations of aggregate-binder can be identified using the BBS test method.

Figure 12 – Pull-out tension values differ for various emulsion types at a range of curing intervals.

Figure 13– Pull-out tension values differ for various substrate types at a range of curing intervals. Table 3 displays experimental results from inter-laboratory testing. Values of the coefficient of variation (COV) are typically less than 10 percent, indicating good repeatability for three replicates in spite of a different substrate preparation procedure. Experimental results are also depicted in Figure 14 for unmodified bitumen and Figure 15 for modified bitumen.

14


Table 3 –BBS test results for materials tested at the University of Stellenbosch. Experimental Factors Binder

Substrate Granite

CRS65 Tillite


Granite CRS65 + 3% LATEX Tillite

Granite SS60 + 3% LATEX Tillite

Cure Interval (hr) 2 6 24 2 6 24 2 6 24 2 6 24 2 6 24 2 6 24 2 6 24 2 6 24

Pull-Out Tensile Strength (kPa) St Average St Error COV Dev 424.6 4.7 2.7 1.1% 533.0 19.5 11.3 3.7% 919.7 46.7 26.9 5.1% 406.6 15.9 9.2 3.9% 765.2 77.2 44.6 10.1% 835.6 57.1 33.0 6.8% 482.4 13.4 7.7 2.8% 466.2 5.6 3.3 1.2% 545.7 106.9 61.7 19.6% 445.4 8.3 4.8 1.9% 459.9 9.8 5.6 2.1% 568.3 51.5 29.7 9.1% 439.1 4.1 2.4 0.9% 824.8 13.9 8.0 1.7% 989.2 169.9 98.1 17.2% 446.3 12.4 7.2 2.8% 846.5 60.4 34.9 7.1% 1095.8 104.8 60.5 9.6% 472.5 6.8 3.9 1.4% 505.9 12.4 7.2 2.5% 477.9 5.6 3.3 1.2% 449.9 6.8 3.9 1.5% 484.2 2.7 1.6 0.6% 717.3

52.8

30.5

7.4%

Figure 14 shows a combination of test results for CRS65 and SS60 bitumen emulsions on granite and tillite aggregates for different loading rates. The following points of interest are noted: • •

•

The CRS65-granite combination outperforms the CRS65-tillite combination at all curing intervals. The pull-out tension appears to increase linearly with curing time. SS60 results show very little strength increase over 24 hours, except on tillite at 24 hours. The result is expected from a slow setting emulsion applied on acidic aggregates such as granite and tillite. These aggregates contain silica and have a strong negative charge in the presence of water. This negative charge attracts positively charged cationic bitumen particles, leading to destabilization of the surfactant system and subsequent coagulation of the bitumen particles. This breaking mechanism is absent when anionic emulsions are used with acidic aggregates (Louw et al,2004) Tests conducted at 900 kPa/s show slight increases in pull–out tensions values over those tested at 700 kPa/s, thereby confirming that higher loading rates result in higher pull-out tension values.

15


Figure 14 – BBS testing of unmodified bitumen shows bond strength development over time.

Figure 15 – BBS testing of modified bitumen also shows bond strength development over time.

16


Figure 15 shows a combination of test results for CRS65 + 3% latex and SS60 + 3% latex on granite and tillite aggregate substrates. The following points of interest are noted: • Tests performed at 950 kPa/s show higher results and confirm the findings of Miller that higher loading rates contribute to higher pull-out tension values. • After 2 hours curing time, results exhibit minimal differences in the results on all four emulsion/aggregate combinations. • After 6 hours the results for CRS65 + 3% latex-granite show higher tensile strength values than SS60 + 3% latex-granite. • After 6 hours, there are limited differences in the results of the CRS65 + 3% latex-tillite and SS60 + 3% latex- tillite. • After 24 hours, the CRS65-granite combination shows significantly higher strength values than the SS60-granite combination but smaller values that the CRS65 + 3% latex-tillite and SS60+ 3 % latex-tillite combinations. • The modified emulsions performs better that the unmodified emulsions after 24 hours. • The CRS65 + 3% latex-tillite and SS60 + 3% latex-tillite combinations values relate well and seem to be superior to granite. 4- CONCLUSION This paper reviewed the development of a new test called the Bitumen Bond Strength (BBS) test and its application to bituminous emulsions. Inter-laboratory test results reinforce the hypothesis that the BBS test protocol (AASHTO TP-91, 2011) can be used to effectively evaluate bond strength of different emulsion types and substrate types. Aside from loading rate, emulsion type and curing interval are identified as the most significant factors contributing to bond strength development. Substrate type is also identified as a significant factor leading to bond strength. Interactions between emulsion type and curing interval are identified as the most significant interaction. Further validation of the BBS test method is needed for the test to be integrated into a performance-based specification system for surface seals. 5- ACKNOWLEDGEMENTS This paper is based on the excellent work done by Mr. Timothy Miller and Prof Hussain Bahia from the University of Wisconsin-Madison. Their research and results forms the basis for this paper. Without their help and permission this paper would not have been possible.

17


6- REFERENCES [1]

Miller, T. “Development of Bond Strength Test for Improved Characterization of Asphalt Emulsions.” Master of Science thesis, University of Wisconsin – Madison, 2010.

[2]

Meng, J. “Affinity of Asphalt to Mineral Aggregate: Pull-Off Test Evaluation.” Master of Science thesis, University of Wisconsin – Madison, 2010.

[3]

Canestrari, F., Cardone, F., Graziani, A., Santagata, F., and Bahia, H. “Adhesive and Cohesive Properties of Asphalt-Aggregate Systems Subjected to Moisture Damage.” Proc., European Asphalt Technology Association, Journal of Road Materials and Pavement Design, 11-32, 2010.

[4]

Bahia, H., Meng, J., Velasquez, R., Miller, T., Daranga, C., and Moraes, R. “Evaluation of the Bitumen Bond Strength (BBS) Test for Moisture Damage Characterization.” Asphalt Research Consortium report prepared for the Federal Highway Administration, University of Wisconsin – Madison, 2010.

[5]

Opus International Consultants. “Performance Based Bitumen Specification.” Discussion paper, New Zealand, 2010.

[6]

Epps, A., Glover, C., and Barcena, R. “A Performance-Graded Binder Specification for Surface Treatments.” Report FHWA/TX-02/1710-1, Texas Department of Transportation, 2001.

[7]

Comité Européen de Normalisation (CEN). “Activity Report of the CEN Ad-Hoc Group Adhesion/Durability.” Report N090, 2009.

[8]

James, A. “Overview of Asphalt Emulsions.” Circular E-C102, Transportation Research Board, Washington, D.C., 2006.

[9]

Redelius, P. and Walter, J. “Emulsion and Emulsion Stability.” Taylor & Francis, New York, 2006.

[10] ASTM. “ASTM D4541-09 Standard Test Method for Pull-Off Strength of Coatings Using Portable Adhesion Testers.” American Society of Testing and Materials, West Conshohocken, PA, 2009. [11] Youtcheff, J., Williams, C. and Stuart, K. “Moisture Sensitivity Testing of Bitumen Using a Pneumatic Adhesion Test.” Eurobitume Workshop, Luxembourg, 1999. [12] Kanitpong, K., Bahia, H. “Role of Adhesion and Thin Film Tackiness of Asphalt Binders in Moisture Damage of HMA.” J. Asphalt Paving Technology, 72, 502-528, 2003. [13] Miller, T., Arega, Z., and Bahia, H. “Correlating Rheological and Bond Properties of Emulsions to Aggregate Retention of Chip Seals.” Proc., Transportation Research Record 1860, Transportation Research Board, Washington, D.C., in press, 2010. [14] Fratta, D. and Daranga, C. “Experimental Validation of the Modified PATTI Test Methodology.” Asphalt Research Consortium report prepared for the Federal Highway Administration, University of Wisconsin – Madison, 2006. [15] ASTM. ASTM D7000-08 Standard test method for sweep test of bituminous emulsion surface treatment samples. American Society for Testing and Materials, West Conshohocken, PA, 2004. [16] Arega, Z. “Rheological Characterization of Asphalt Emulsions for Chip Seal Applications.” Master of Science thesis, University of Wisconsin – Madison, 2009.

18


[17] Louw, K., Spence, K. and Kuun, P. “The use of bitumen emulsions as a cost effective solution for constructing seals during winter.” Proc., 8th Conference on Asphalt Pavements for Southern Africa (CAPSA), Sun City, South Africa, 2004.

7- AUTHOR BIOGRAPHIES André Greyling is a Professionally Registered Civil Engineer and a Director at BVi Consulting Engineers in Cape Town, South Africa. He is currently completing a part-time Msc. Eng degree in Pavement Engineering under the guidance of Professor Kim Jenkins at the University of Stellenbosch. Timothy Miller is a research assistant and teaching assistant in the Department of Civil and Environmental Engineering at the University of Wisconsin – Madison. Mr. Miller is pursuing his Ph.D. degree at UW-Madison and leading efforts to improve frameworks for analyzing asphalt pavement sustainability. Dr. Hussain Bahia is a professor of Civil and Environmental Engineering at the University of Wisconsin-Madison. Prior to joining the University of Wisconsin in 1996, he served as a research assistant, research associate, and Assistant Professor at Pennsylvania State University. Dr. Bahia has served as a principal investigator and co-investigator for a number of research projects and is the author/co-author of many research papers related to asphalt binders, mixtures, and pavement construction. Dr. Kim Jenkins studied Civil Engineering at the University of Natal in South Africa and worked for 12 years in the consulting engineering environments. He focused on geotechnics, road materials and pavement design, specializing in road rehabilitation and recycling technology. Since 1996 he has worked at Stellenbosch University and is currently the incumbent SANRAL Chair in Pavement Engineering.

19