Citation
Evaluation of crack resistant concrete for Colorado bridge decks

Material Information

Title:
Evaluation of crack resistant concrete for Colorado bridge decks
Creator:
Cavaliero, Robert Wayne
Publication Date:
Language:
English
Physical Description:
xix, 244 leaves : illustrations ; 28 cm

Subjects

Subjects / Keywords:
Bridges -- Floors ( lcsh )
Concrete bridges -- Design and construction ( lcsh )
Concrete bridges -- Cracking ( lcsh )
Bridges -- Floors ( fast )
Concrete bridges -- Cracking ( fast )
Concrete bridges -- Design and construction ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 203-205).
Statement of Responsibility:
by Robert Wayne Cavaliero.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
519547259 ( OCLC )
ocn519547259
Classification:
LD1193.E53 2009m C38 ( lcc )

Downloads

This item has the following downloads:


Full Text
EVALUATION OF CRACK RESISTANT
CONCRETE FOR COLORADO BRIDGE DECKS
by
Robert Wayne Cavaliero
B.S., Civil Engineering, University of Colorado at Denver. 2007
A thesis submitted to the
University of Colorado at Denver/Health Sciences Center in partial fulfillment of the requirements for the degree of Master of Science, Structural Engineering Civil Engineering


2009
This thesis for the Master of Science degree by
Robert W. Cavaliero has been approved
by

Date
Cheng Yu Li


Cavaliero, Robert Wayne (MS, Structural, Civil Engineering Department) Evaluation of Crack Resistant Concrete for Colorado Bridge Decks Thesis directed by Dr. Stephan A. Durham
ABSTRACT
Cracking of reinforced concrete is a problem that has baffled maintenance and bridge engineers since it was first developed. Cracking allows water and contaminates to enter the structure and corrode the reinforcing steel. A concrete structure that is constructed with a concrete that is less susceptible to cracking will last longer and be more maintenance free than one that is susceptible to cracking. The Colorado Department of Transportation (CDOT) has developed a Class H and HT concrete that is meant to fill the requirement of a crack resistant concrete. Though the current specification is an improvement over the Class D concrete that is commonly used there are still some problems with cracking.
The basis for this research includes the design of over ten concrete mixtures in an effort to create a more crack resistant concrete than the current CDOT Class IT and HT concrete specification. Cracking is known to be the result of many factors including shrinkage, which contributes in large amounts. The concrete mixtures designed for this research were designed with water to cementitious materiaTs amounts and cement replacement percentages both above and below the current


specifications. The design approach was intended to investigate the effect of individual and multiple cementitious materials replacement on several tests of fresh and hardened concrete; restrained and unrestrained shrinkage strain, compressive strength, rate of strength gain, freeze/thaw durability, permeability.
A national survey was conducted and offered to all fifty, state DOT bridge and/or materials engineers. They were asked about current and past research involving crack resistant concrete as well as comments regarding their state DOT specifications currently used for bridge decks. The results of this survey are used to provide valuable feedback to the Research Team for use in improving the current CDOT Class H and HT specifications. The information was also taken into consideration during the design process of the concrete mixtures for this research.
A more crack resistant concrete mixture was developed through this study which was designed for better performance in Colorados bridge decks, including less maintenance and improved durability over the life of the structure.
This abstract accurately represents the content of the candidates thesis. I recommend its publication.
Signed
Stephan A. Durham


DEDICATION PAGE
I dedicate this thesis to my friends and family for their patience and understanding throughout this process. You can never understand how much I appreciate your understanding for the distance created in our relationships and your patience in waiting for me to complete such a project. To my parents for their never-ending support and inspiration to become what I never thought was possible.


ACKNOWLEDGMENT
I sincerely thank my academic advisor, Dr. Stephan Durham, who secured this research for the benefit of both the University of Colorado at Denver and the Colorado Department of Transportation. His continued advisement has taught me and so many others. Also, I would like to thank Dr. Kevin Rens and Dr. Li for participating on my thesis committee.
Thank you to the Colorado Department of Transportation for a great experience. My sincere Thanks to Rui Liu, Tom Thuis, Randy Ray, Driss Majdoub, Adam Kardos, Logan Young and the many others who contributed to this work.
I would like to thank Mr. Ken Stevens with Campbell Scientific Laboratories for his many hours of help fine-tuning the data logger program.
Thank you to all those who donated time and materials to complete this research; Dan Bentz and Jesse (BestWay Aggregate), Matt Finger and Kevin Kane (Holcim Cement), Brandon Cooke (BASF), Tom Green and Bill Hart (W.R.Grace), George Grygar and Miles Dee (txi), David Darwin and JoAnne Browning (University of Kansas), and Jason Weiss (Purdue University). Additionally, I would like to thank the faculty and staff of the University of Colorado at Denver, Civil Engineering Department for their support and guidance throughout my educational career at
UCD.


TABLE OF CONTENTS
Figures..................................................................xii
Tables...................................................................xviii
Chapter
1. Introduction........................................................1
1.1 Concrete............................................................1
1.1.1 Problematic Cracking in Concrete Bridges............................1
2. Background..........................................................4
2.1 Colorado Department of Transportation...............................4
2.1.1 Research I nterest..................................................4
2.1.2 Current Specifications..............................................6
2.1.2.1 Class H Specifications..............................................6
2.1.2.2 Class HT Specifications.............................................7
2.2 Cracking in Concrete................................................8
2.2.1 Importance of Cracking in Concrete..................................8
2.3 Causes of Cracking in Concrete......................................9
2.3.1 Internal Stresses...................................................9
2.3.2 External Stresses and Normal Use Degradation.......................10
2.3.3 Restraint..........................................................10
2.3.4 Shrinkage Strain...................................................10
2.4 Research Objectives................................................12
2.4.1 Objectives of Investigation........................................12
3. Literature Review..................................................13
3.1 Preface ...........................................................13
3.2 Curing ............................................................13
3.3 Concrete Shrinkage.................................................14
3.3.1 Effect of Restraint on Shrinkage...................................14
3.3.2 Effect of Curing on Shrinkage .....................................15
3.3.2.1 Internal Curing Using Light Weight Aggregate.......................20
3.4 Design Mixture Factors Affecting Cracking in Concrete..............23
3.4.1 Silica Fume........................................................23
3.4.2 Fly Ash ...........................................................25
3.4.3 Water to Cementitious Materials Ratio (w/cm).......................28
3.4.4 Cement Content.....................................................29
3.4.5 Cement Type........................................................30
vii


30
32
33
35
37
37
37
39
42
44
46
47
49
49
52
52
52
53
54
54
55
55
55
56
56
57
57
58
58
58
60
60
60
60
61
62
General Effects of Cement Fineness.................
Coarse-Ground Cement...............................
Shrinkage Compensating Cements.....................
Aggregate Content..................................
Aggregate Composition..............................
General ...........................................
Aggregate Composition and Water to Cementitious
Materials Content..................................
Unrestrained Shrinkage Test........................
AASHTO PP34 / ASTM C 1581..........................
Restrained Ring Shrinkage Test.....................
Length Change......................................
Admixtures.........................................
Problem Statement..................................
Statement..........................................
State DOT Survey...................................
General ...........................................
Survey Response....................................
State DOT Bridge Deck Cracking Problem.............
Potential Causes for Bridge Deck Cracking..........
Rate of Concrete Strength Gain.....................
AASHTO PP34 Ring Test Usage by State DOTs..........
Mixture Design Issues..............................
Mixture Design Modifications Used to Improve Concrete
Performance........................................
Shrinkage-Reducing Admixtures......................
Shrinkage Compensating Cement......................
Factors Affecting Cracking (Mixture Design)........
Beneficial Factors that Reduce Concrete Cracking...
Water-to-Cementitious Materials Ratio..............
Curing Practices...................................
DOT Survey Conclusion..............................
Experimental Design................................
Design Plan........................................
Literature Review..................................
Mixture Design Process.............................
Mixture Designs....................................
Cement Type........................................
viii


6.1.3.2 Supplementary Cementitious Materials.............................63
6.1.3.3 Chemical Admixtures..............................................63
6.1.3.4 Aggregate Type...................................................64
6.2 Acquisition of Raw Materials.....................................65
6.2.1 Cement ..........................................................65
6.2.2 Aggregate........................................................67
6.2.3 Admixtures.......................................................68
6.2.3.1 High-Range Water Reducing Admixture (H.R.W.R.A.).................68
6.2.3.2 Air-Entraining Agent (A.E.A.)....................................69
6.2.3.3 Shrinkage-Reducing Admixture (S.R.A.)............................69
6.2.3.4 Set Retarder (RET)...............................................69
6.3 Testing .........................................................69
6.4 Data Analysis....................................................71
7. Experimental Results.............................................72
7.1 Problems with this Study.........................................72
7.1.1 Fabrication of Steel Rings.......................................72
7.1.2 Data Acquisition System..........................................72
7.1.3 Re-Batch of Mixture #1 (0.38-6.8-FA20-SF5-II) and
Mixture #2 (0.42-6.2-FA16-SF3.5-II).............................73
7.2 Fresh Concrete Properties........................................75
7.2.1 Slump ...........................................................75
7.2.1.1 Cement Type......................................................76
7.2.1.2 Supplementary Cementitious Materials.............................77
7.2.1.3 Chemical Admixtures..............................................77
7.2.1.4 Aggregate Type...................................................78
7.2.2 Air Content......................................................79
7.2.3 Unit Weight......................................................81
7.2.4 Concrete Temperature.............................................83
7.4 Hardened Concrete Tests..........................................85
7.4.1 Compressive Strength.............................................85
7.4.1.1 Mixtures Having Inadequate 56-Day Strength.......................87
7.4.1.2 Normalization of Compressive Strength............................89
7.4.1.3 Comparison of Mixture #1 (0.38-6.8-FA20-SF5-II) and
Mixture #2 (0.42-6.2-FA 16-SF3.5-II), Batch One and Two.........92
7.4.1.4 Early-Age Compressive Strength...................................96
7.4.1.4.1 Cement Type....................................................98
7.4.1.4.2 Supplementary Cementitious Materials..........................102
7.4.1.4.3 Chemical Admixtures...........................................106
7.4.1.4.4 Aggregate Type................................................107
7.4.1.5 Ultimate Strength (28-day and 56-Day)...........................109
IX


7.4.1.5.1 Cement Type......................................................109
7.4.1.5.2 Supplementary Cementitious Materials.............................112
7.4.1.5.3 Chemical Admixtures..............................................114
7.4.1.5.4 Aggregate Type..................................................115
7.4.2 Permeability......................................................117
7.4.2.1 General ..........................................................117
7.4.2.2 Rapid Chloride Ion Penetrability Test............................119
7.4.2.2.1 Cement Type......................................................122
7.4.2.2.2 Supplementary Cementitious Materials.............................127
7.4.2.2.3 Chemical Admixtures..............................................130
7.4.2.2.4 Aggregate Type...................................................132
7.4.3 Durability........................................................134
7.4.3.1 General ..........................................................134
7.4.3.2 Durability Analysis...............................................155
7.4.3.2.1 Cement Type......................................................155
7.4.3.2.2 Supplementary Cementitious Materials.............................159
7.4.3.2.3 Chemical Admixtures..............................................160
7.4.3.2.4 Aggregate Type...................................................161
7.4.4 Restrained Shrinkage Strain......................................162
7.4.4.1 General ..........................................................162
7.4.4.2 Strain Analysis...................................................166
7.4.4.2.1 Cement Type......................................................166
7.4.4.2.2 Supplementary Cementitious Materials.............................175
7.4.4.2.3 Chemical Admixtures..............................................178
7.4.4.2.4 Aggregate Type...................................................183
7.4.4.3 Paste Content (Volume)............................................189
8. Conclusions and Recommendations.........................................196
8.1 Fresh Concrete Properties.........................................196
8.1.1 Slump ............................................................196
8.1.2 Air Content.......................................................197
8.1.3 Unit Weight.......................................................197
8.1.4 Temperature.......................................................198
8.2 Mixture Design Properties.........................................198
8.2.1 General ..........................................................198
8.3 Recommendations...................................................202
x


Appendix
A. Concrete Design Mixtures...........................................206
B. Materials Product Data.............................................217
C. DOT Survey.........................................................234
D. Photographs of Cracked Restrained Ring Shrinkage Test Specimens...239
Bibliography ..........................................................203


FIGURES
Figure
5.1 DOT Respondents Map..............................................53
7.1 Slump Test Results, (ASTM C 143, AASHTO T 119)...................78
7.2 Air Content, (ASTM C 231, AASHTO T 152)..........................81
7.3 Unit Weight (ASTM C 138, AASHTO T 121) vs. Air
Content (ASTM C 231, AASHTO T 152................................83
7.4 Concrete Temperature, (ASTM C 1064, AASHTO T 309)................84
7.5 Photograph of Compressive Strength Failure
(ASTM C 39, AASHTO T 22).........................................86
7.6 56-Day Compressive Strength (ASTM C 39, AASHTO T 22).............89
7.7 56-Day Compressive Strength vs. 56-Day Compressive Strength (Normalized for Air Content), (ASTM C
39, AASHTO T 22).................................................91
7.8 28-Day Compressive Strength, CDOT Control Mixture #1 (0.38-6.8-FA20-SF5-1I), Batch One vs. Batch Two,
(ASTM C 39, AASHTO T 22).........................................93
7.9 28-Day Compressive Strength, CDOT Control Mixture #2 (0.42-6.2-FA16-SF3.5-II). Batch One vs. Batch Two, (ASTM
C 39, AASHTO T 22)...............................................94
7.10 Early-Age Compressive Strength, (ASTM C 39, AASHTO T 22)........97
7.11 Early-Age Compressive Strength, CDOT Control Mixture #1 (0.38-6.8-FA20-SF5-II) (Type II Cement) and Mixture #3 (0.38-6.8-FA20-SF5-G) (Type G,
Coarse-Ground Cement), (ASTM C 39, AASHTO T 22)..................99
7.12 Early-Age Compressive Strength, CDOT Control Mixture #2 (0.42-6.2-FA16-SF3.5-II) (Type II Cement) and Mixture #4 (0.42-6.2-FA16-SF3.5-G) (Type G, Coarse-Ground Cement),
(ASTM C 39, AASHTO T 22).........................................100
7.13 Early-Age Compressive Strength, Mixture #5 (0.44-6.5-FA30-II), Mixture #6 (0.44-6.5-FA30-SF5-II), and
Mixture #7 (0.44-6.5-BFS50-II),(ASTM C 39, AASHTO T 22)..........104
7.14 Early-Age Compressive Strength, Mixture #8 (0.44-6.0-FA30-SRA-II) (Shrinkage Reducing Admixture) and Mixture #9 (0.44-6.0-FA30-RET-II)
(Set Retarding Admixture), (ASTM C 39, AASHTO T 22)..............106
7.15 Early-Age Compressive Strength, Mixture #10 (0.42-6.0-II-Light Weight Aggregate) and Mixture #11 (0.42-6.0-II-Normal
Weight Aggregate), (ASTM C 39, AASHTO T 22)......................108
xii


7.16 Compressive Strength, CDOT Control Mixture #1 (0.38-6.8-FA20-SF5-II) (Type II Cement) and Mixture
#3 (0.38-6.8-FA20-SF5-G) (Type G, Coarse-Ground Cement)
(ASTM C 39, AASHTO T 22).........................................110
7.17 Compressive Strength, CDOT Control Mixture #2
(Type 11 Cement) and Mixture #4 (0.42-6.2-FA16-SF3.5-G)
(Type G, Coarse-Ground Cement), (ASTM C 39, AASHTO T 22).........111
7.18 Compressive Strength, Mixture #5 (0.44-6.5-FA30-II), Mixture #6 (0.44-6.5-FA30-SF5-II), and Mixture #7 (0.44-6.5-BFS50-II),
(ASTM C 39, AASHTO T 22).........................................113
7.19 Compressive Strength, Mixture #8 (0.44-6.0-FA30-SRA-II)
(Shrinkage Reducing Admixture) and Mixture #9 (0.44-6.0-FA30-RET-II) (Set Retarding Admixture), (ASTM
C 39, AASHTO T 22)...............................................114
7.20 Compressive Strength, Mixture #10 (0.42-6.0-II-Light Weight Aggregate) and Mixture #11 (0.42-6.0-II-Normal Weight
Aggregate), (ASTM C 39, AASHTO T 22).............................116
7.21 Photograph of R.C.l.P. Test Setup...............................118
7.22 Rapid Chloride Ion Penetrability Test Results (ASTM C 1202,
AASHTO T 227).................................................... 121
7.23 56-Day Rapid Chloride Ion Penetrability Test Results (ASTM
C 1202, AASHTO T 227)............................................ 122
7.24 28-Day and 56-Day Rapid Chloride-Ion Penetrability Test Results,
CDOT Control Mixture #1 (0.38-6.8-FA20-SF5-II)
(Type II Cement) and Mixture #3 (0.38-6.8-FA20-SF5-G)
(Type G, Coarse-Ground Cement), (Permeability, ASTM C 1202, AASHTO T 227)...................................... 124
7.25 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results, CDOT Control Mixture #2 (0.42-6.2-FA 16-SF3.5-II)
(Type II Cement) and Mixture #4 (0.42-6.2-FA16-SF3.5-G)
(Type G, Coarse-Ground Cement),(ASTM C 1202, AASHTO T 227)....... 126
7.26 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results, Mixture #5 (0.44-6.5-FA30-II), Mixture #6 (0.44-6.5-FA30-SF5-II), and Mixture #7 (0.44-6.5-BFS50-II),
(Permeability, ASTM C 1202, AASHTO T 227)........................ 129
7.27 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results,
Mixture #5 (0.44-6.5-FA30-II), Mixture #6 (0.44-6.5-FA30-SF5-II), and Mixture #7 (0.44-6.5-BFS50-II), (Permeability, ASTM
C 1202, AASHTO T 227)............................................ 130
xiii


7.28 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results,
Mixture #8 (0.44-6.0-FA30-SRA-II) (Shrinkage Reducing Admixture) and Mixture #9 (0.44-6.0-FA30-RET-II) (Set Retarding Admixture), (Permeability, ASTM C 1202, AASHTO T 227)................... 131
7.29 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results,
Mixture #8 (0.44-6.0-FA30-SRA-II) (Shrinkage Reducing Admixture) and Mixture #9 (0.44-6.0-FA30-RET-II) (Set Retarding Admixture), (Permeability, ASTM C 1202, AASHTO T 227)).................. 132
7.30 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results,
Mixture #10 (0.42-6.0-II-Light Weight Aggregate) and Mixture #11 (0.42-6.0-11-Normal Weight Aggregate) (Permeability, ASTM C 1202, AASHTO T 227)............................................... 134
7.31 Photograph of Freeze/Thaw Chamber (ASTM C 666, Procedure A)......136
7.32 Photograph of Durability Testing Apparatus (ASTM C
666, Procedure A)..................................................137
7.33 Photograph of Durability Testing Apparatus (ASTM C 666,
Procedure A).......................................................138
7.34 Durability Factor and Air Content, CDOT Control Mixture #1 (0.38-6.8-FA20-SF5-II) and CDOT Control Mixture #2
(0.42-6.2-FA 16-SF3.5-II)..........................................156
7.35 Durability Factor and Air Content, CDOT Control Mixture #1
(0.38-6.8-FA20-SF5-II) and Mixture #3 (0.38-6.8-FA20-SF5-G)........157
7.36 Durability Factor and Air Content, CDOT Control Mixture #2
(0.42/6.2/FA 16/SF3.5/II) and Mixture #4 (0.42/6.2/FA16/SF3.5/G)...158
7.37 Durability Factor and Air Content, Mixture #5 (0.44/6.5/FA30/II),
Mixture #6 (0.44/6.5/FA30/SF5/II), and Mixture #7
(0.44/6.5/BFS50/II)................................................160
7.38 Durability Factor and Air Content, Mixture #8
(0.44-6.0-FA30-SRA-II).............................................162
7.39 Photograph of Restrained Ring Shrinkage Specimen (ASTM
C 1581, AASHTO PP34)...............................................163
7.40 Photograph of Restrained Ring Shrinkage Specimen, (ASTM
C 1581, AASHTO PP34)...............................................165
7.41 Restrained Shrinkage Strain, CDOT Control Mixture #1 (0.38-6.8-FA20-SF5-II) and Mixture #3 (0.38-6.8-FA20-SF5- G),
(ASTM C 1581, AASHTO PP34).........................................167
7.42 % of 56-Day Strength Achieved at Respective Age, Mixture #1 (0.38-6.8-FA20-SF5-II) and Mixture #3 (0.38-6.8-FA20-
SF5-G), (ASTM C 39, AASHTO T 22)...................................169
XIV


7.43 % of Ultimate Strain Achieved, CDOT Control Mixture #1 (0.38-6.8-FA20-SF5-1I) and Mixture #3 (0.38-6.8-FA20-
SF5-G), (ASTM C 1581, AASHTO PP34)..............................170
7.44 % of 56-Day Strength Achieved at Respective Age,
CDOT Control Mixture #2 (0.42/6.2/FA16/SF3.5/11) and Mixture #4 (0.42/6.2/FA16/SF3.5/G), (ASTM C 39,
AASHTO T 22)....................................................171
7.45 % of Ultimate Strain Achieved at Respective Age, CDOT
Control Mixture #2 (0.42/6.2/FA16/SF3.5/II) and Mixture #4 (0.42/6.2/FA16/SF3.5/G), (ASTM C 1581, AASHTO PP34).........173
7.46 Restrained Shrinkage Strain, CDOT Control Mixture #2 (0.42/6.2/FA16/SF3.5/II) and Mixture #4 (0.42/6.2/FA16/SF3.5/G),
(ASTM C 1581, AASHTO PP34).......................................174
7.47 Restrained Shrinkage Strain, Mixture #5 (0.44/6.5/FA30/II),
Mixture #6 (0.44/6.5/FA30/SF5/II), and Mixture #7
(0.44/6.5/BFS50/II), (ASTM C 1581, AASHTO PP34)..................176
7.48 % of Ultimate Strain Achieved at Respective Age, Mixture #5 (0.44/6.5/FA30/II), Mixture #6 (0.44/6.5/FA30/SF5/II), and Mixture #7 (0.44/6.5/BFS50/II), (ASTM C 1581,
AASHTO PP34).....................................................178
7.49 Restrained Shrinkage Strain, Mixture #8 (0.44-6.0-FA30-SRA-II) and Mixture #9 (0.44-6.0-FA30-RET-II), (ASTM C 1581,
AASHTO PP34).....................................................180
7.50 % of 56-Day Strength Achieved, Mixture #8 (0.44-6.0-FA30-SRA-II) and Mixture #9 (0.44-6.0-FA30-RET-II),
(ASTM C 39, AASHTO T 22).........................................181
7.51 % of Ultimate Strain Achieved, Mixture #8 (0.44-6.0-FA30-SRA-II) and Mixture #9 (0.44-6.0-FA30-RET-II), (ASTM C 1581,
AASHTO PP34).....................................................183
7.52 % of 28-Day Strength Achieved, Mixture #10 (0.42-6.0-I1-L.W.A) and Mixture #11 (0.42-6.0-II-Norm.Wt.), (ASTM C 39,
AASHTO T 22).....................................................185
7.53 % of Ultimate Strain Achieved, Mixture #10 (0.42-6.0-II-L.W.A) and Mixture #11 (0.42-6.0-II-Norm.Wt.), (ASTM C 1581,
AASHTO PP34).....................................................187
7.54 Restrained Shrinkage Strain, Mixture #10 (0.42-6.0-II-L.W.A) and Mixture #11 (0.42-6.0-II-Norm.Wt.), (ASTM C
1581, AASHTO PP34)...............................................188
xv


7.55 % of Ultimate Strain Achieved vs. Paste Content (29 vs. 25%), Mixture 5 (0.44/6.5/FA30/II) and Mixture #6 (0.44/6.5/FA30/SF5/II) vs. Mixture #8 (0.44/6.0/FA30/SRA/II)
and Mixture #9 (0.44/6.0/FA30/RET/II) (ASTM C 1581,
AASHTO PP34)..................................................193
7.56 % Ultimate Strain Achieved vs. Paste Content (28 vs. 26%),
Mixture #1 (0.38/6.8/FA20/SF5/II) and Mixture #2 (0.42/6.2/FA16/SF3.5/II) respectively, (ASTM C 1581,
AASHTO PP34)..................................................195
Appendix B
Figure
B.l Fine Aggregate Gradation (ASTM C 33).............................217
B.2 Coarse Aggregate Gradation (ASTM C 33)..........................218
B.3 WesTest Aggregate Test Results....................................219
B.4 Holcim Type II Cement Properties..................................222
B.5 Boral Material Technologies, Class F Fly Ash (ASTM
C 618 T Report)...................................................223
B.6 W.R.Grace, Daracem 19, High Range Water Reducing
Admixture,Product Data............................................226
B.7 W.R.Grace, Daravair AT60, Air-Entraining Admixture
(ASTM C 260), Product Data........................................228
B.8 BASF Tetraguard AS20, Shrinkage-Reducing Admixture,
Product Data......................................................230
B.9 BASF Pozzolith 100XR, Set-Retarder Admixture,
Product Data......................................................232
Appendix C
Figure
C. 1 DOT Survey, Questions #1 and #2.................................234
C.2 DOT Survey, Questions #3, #4, #5, and #6........................235
C.3 DOT Survey, Questions #7, #8, #9, and #10.......................236
C.4 DOT Survey, Questions #11, #12, and #13.........................237
C.5 DOT Survey, Respondent Contact Information..........................238
XVI


Figure
Appendix D
D.l Photograph of Mixture #3 (0.38-6.8-FA20-SF5-G), Ringl
Restrained Ring Shrinkage Test Specimen............................239
D.2 Photograph of Mixture #2 (0.42-6.2-FA 16-SF3.5-II), Ring2
Restrained Ring Shrinkage Test Specimen............................240
D.3 Photograph of Mixture #4 (0.42-6.2-FA 16-SF3.5-G), Ring2
Restrained Ring Shrinkage Test Specimen............................241
D.4 Photograph of Mixture #5 (0.44-6.5-FA30-II), Ringl
Restrained Ring Shrinkage Test Specimen............................242
D.5 Photograph of Mixture #6 (0.44-6.5-FA30-SF5-II), Ring2
Restrained Ring Shrinkage Test Specimen............................243
D.6 Photograph of Mixture #9 (0.44-6.0-FA30-RET-II), Ringl
Restrained Ring Shrinkage Test Specimen............................244
xvii


TABLES
Table
2.1 Class H and Class HT mixture specifications.............................4
6.1 Mixture Design Matrix..................................................61
6.2 Class G Oilwell Cement Compounds.......................................65
6.3 Class G Oilwell Cement Chemical and Physical Properties................66
6.4 Class G Oilwell Cement Compressive Strength Properties.................66
6.5 Holcim Type II Cement Chemical and Physical Properties.................67
6.6 Holcim Type II Cement Compressive Strength Properties..................67
6.7 Fresh and Hardened Concrete Properties Tests...........................71
7.1 Fresh Concrete Properties..............................................75
7.2 Compressive Strength (ASTM C 39, AASHTO T 22)..........................87
7.3 Normalized Compressive Strength........................................90
7.4 Permeability Rating per Coulombs Passed...............................119
7.5 Rapid Chloride Ion Penetrability Results (ASTM C 1202,
AASHTO T 227)....................................................... 120
7.6 Mixture #1 (0.38-6.8-FA20-SF5-II), freeze/thaw results................139
7.7 Mixture #2 (0.42-6.2-FA16-SF3.5-II), freeze/thaw results..............140
7.8 Mixture #3 (0.38-6.8-FA20-SF5-G), freeze/thaw results.................141
7.9 Mixture #4 (0.42-6.2-FA 16-SF3.5-G), freeze/thaw results..............141
7.10 Mixture #5 (0.44-6.5-FA30-II), freeze/thaw results...................142
7.11 Mixture #6 (0.44-6.5-FA30-SF5-II), freeze/thaw results...............142
7.12 Mixture #7 (0.44-6.5-BFS50-II), freeze/thaw results..................143
7.13 Mixture #8 (0.44-6.0-FA30-SRA-II), freeze/thaw results...............143
7.14 Mixture #9 (0.44-6.0-FA30-RET-II), freeze/thaw results...............144
7.15 Mixture #10 (0.42-6.0-II-light weight aggregate), freeze/thaw result.144
7.16 Mixture #11 (0.42-6.0-II-normal weight aggregate), freeze/thaw results.. 144
7.17 Mixture #1 (0.38-6.8-FA20-SF5-II), relative modulus of elasticity....147
7.18 Mixture #2 (0.42-6.2-FA16-SF3.5-II), relative modulus of elasticity..148
7.19 Mixture #3 (0.38-6.8-FA20-SF5-G), relative modulus of elasticity.....149
7.20 Mixture #4 (0.42-6.2-FA 16-SF3.5-G), relative modulus of elasticity..150
7.21 Mixture #5 (0.44-6.5-FA30-II), relative modulus of elasticity........150
7.22 Mixture #6 (0.44-6.5-FA30-SF5-II), relative modulus of elasticity....151
7.23 Mixture #7 (0.44-6.5-BFS50-II), relative modulus of elasticity.......151
7.24 Mixture #8 (0.44-6.0-FA30-SRA-II), relative modulus of elasticity....152
7.25 Mixture #9 (0.44-6.0-FA30-RET-II), relative modulus of elasticity....152
7.26 Mixture #10 (0.42-6.0-II-light weight aggregate), relative
modulus of elasticity...............................................153
xviii


7.27 Mixture #11 (0.42-6.0-II-normal weight aggregate), relative
modulus of elasticity..................................................153
7.28 Durability Factors.....................................................154
7.29 Design Mixture Properties..............................................190
8.1 Compressive Strength, Permeability, and Restrained
Shrinkage Test Results.................................................200
8.2 Comparison Between Study Mixtures and Class H and HT
Specification Requirements.............................................201
Appendix A
Figure
A.l Concrete Design Spreadsheet, Mixture #1 (0.38-6.8-FA20-SF5-II)....206
A.2 Concrete Design Spreadsheet, Mixture #2 (0.42/6.2/FA16/SF3.5/11) .207
A.3 Concrete Design Spreadsheet, Mixture #3 (0.38-6.8-FA20-SF5-G).....208
A.4 Concrete Design Spreadsheet, Mixture #4 (0.42/6.2/FA16/SF3.5/G)...209
A.5 Concrete Design Spreadsheet, Mixture #5 (0.44/6.5/FA30/II)........210
A.6 Concrete Design Spreadsheet, Mixture #6 (0.44/6.5/FA30/SF5/II)....211
A.7 Concrete Design Spreadsheet, Mixture #7 (0.44/6.5/BFS50/II).......212
A.8 Concrete Design Spreadsheet, Mixture #8 (0.44-6.0-FA30-SRA-II)....213
A.9 Concrete Design Spreadsheet, Mixture #9 (0.44-6.0-FA30-RET-I1)....214
A. 10 Concrete Design Spreadsheet, Mixture #10 (0.42-6.0-II-L.W.A))......215
A.l 1 Concrete Design Spreadsheet, Mixture #11 (0.42-6.0-II-Norm.Wt.)....216
XIX


Chapter 1
Introduction
1.1 Concrete
1.1.1 Problematic Cracking in Concrete
Within the past five years, the Colorado Department of Transportation (CDOT) has experienced a continued problem with cracking of bridge decks. In 2003, CDOT implemented concrete mixture designs Class H and Class HT into the CDOT Standard Specification for Road and Bridge Construction. Class H and HT were developed to provide crack resistant concrete structures and were intended to be used in the construction of bridges and other concrete structures (CDOT, 2005). Recently, the CDOT has noticed cracking in several bridge decks using these concrete specifications.
Cracking in residential or city roadways will allow for delayed repair. Traffic can swerve to avoid cracks or potholes for months if they get large enough. However, bridge decks are suspended above rivers, mountain valleys, deep crevasses, and sometimes, other roadways carrying the traveling public. If a bridge deck develops sufficient cracking as to damage its structural integrity a failure could result and be disastrous. The result of such large structural failures is of immense possibilities, most of which result in human injury or death. In addition
1


to safety, CDOT and other state DOTs are interested in low-cracking potential concrete in an effort to reduce maintenance costs and delays to the motoring public. Ultimately, the primary objective is to improve the performance of concrete bridge decks in Colorado by minimizing cracking potential of the concrete mixtures used in them.
Cracking in reinforced concrete structures allows water and contaminants to migrate inside the structure where it can cause deterioration of the reinforcing steel as well as the surrounding concrete matrix. Water that is able to penetrate through the bridge superstructure can also cause damage to the substructure and affect bridge aesthetics. The de-icing chemicals used during inclement weather to provide safe driving conditions in combination with air and water accelerates the corrosion of reinforcing steel (rust or oxidation). The existing bond between the concrete and the steel diminishes as the corrosion process progresses, jeopardizing the integrity of the structure. When a bridge is in service and experiences cracking, naturally the cracks grow with time. This allows for more water and de-icing chemicals to enter the deck and degrade the reinforcing steel, creating the need for replacement or repair earlier than normal. This perpetuation of bridge deterioration requires costly and labor-intensive repair.
To minimize the amount of cracking and reduce maintenance costs, Class H and HT concrete mixtures were analyzed in this Thesis to ensure the concrete meets the
2


expectations of the CDOT. Further, additional mixtures were evaluated for their effectiveness in reducing cracking in concrete structures. To accomplish this, eleven concrete mixtures were designed with low-cracking potential as the primary objective. The results of this study and recommendations to the CDOT are included in this thesis.
3


Chapter 2
Background
2.1 Colorado Department of Transportation
2.1.1 Research Interest
In 2003, the Colorado Department of Transportation (CDOT) revised their Standard Specifications for Road and Bridge Construction to include two new classes of structural concrete. Class H and Class HT concrete were included into the standard specifications as a crack resistant concrete. These concretes are currently used in the construction of bridges and other concrete structures. Class H concrete is used for concrete bridge decks without a topping slab and waterproofing membrane [Xi et. al, 2003], Class HT concrete is used as a top layer for exposed concrete bridge decks. The design criterion for each of these concrete classes is shown below.
Table 2.1 Class H and Class HT mixture specifications
Material Class H Class HT
Cement [Cl 450 500 lbs/yd3 450 500 lbs/yd3
Fly Ash [FA1 90- 125 lbs/yd3 90- 125 lbs/yd3
Silica Fume [SF] 20 30 lbs/yd3 20 30 lbs/yd3
C + FA + SF 580 640 lbs/yd3 580 640 lbs/yd3
Course Aggregate AASHTO M 43 Size No. 67> 55% AASHTO M 43 Size No. 7 of 8 > 55%
4


A study on Colorado bridge decks was published in March 2003 [Xi et al, 2003]. The objectives of this study were twofold. First, the extent and causes for bridge deck cracking was investigated. Secondly, concrete material properties, construction practices, and design specifications where examined as to possible causes for bridge deck cracking. A literature review within this study concluded that cracking in early age bridge decks is a result of material, design, construction, and environment. High early age shrinkage was found to be a major cause for this cracking problem. In addition, the structural design had a direct role in cracking as well. Cracks were typically noticed above girders and piers. Placement and curing can have a significant role in cracking, primarily plastic shrinkage cracking. Recommendations regarding materials, design factors, and construction practices were included in the Final Report. Cement and silica fume content, water/cement ratio, and the rate of strength gain were key recommendations regarding materials included in the report.
Recently, the CDOT has discovered a number of bridge decks throughout the state constructed with Class H and Class HT concrete that exhibit cracking. It is suspected that the rate of strength gain for these concrete mixtures may in part be a contributing factor to this cracking. Several bridge decks have obtained the 28-day compressive strength within three days. Other factors that influence cracking
5


include: types and amount of aggregate, cement content and type, water/cement ratio, and air content. These are discussed in more detail in Chapter 3 of this thesis.
Colorados harsh weather conditions make it essential for the states bridge decks to have strict performance and mixture specifications. Early age cracking of bridge decks can decrease the life of the structure and increase maintenance costs immensely. Traffic safety and an efficient use of materials and labor are of great interest to the CDOT. For this reason, the CDOT has requested an investigation into developing a new specification for concrete with low cracking potential.
2.1.2 Current Specifications
2.1.2.1 Class H Specifications
Class H concrete is used for bare concrete bridge decks with no waterproofing membrane. Below is a summary of current CDOT Class H and HT specifications.
56-day compressive strength of 4500 lbs./in.2;
Required air content of 5% 8% are required;
Water-to-Cementitious Ratio (w/cm) ranging from 0.38 0.42;
An approved water reducing admixture ;
A minimum of 55 percent AASHTO M 43 size No. 67 coarse aggregate by weight of total aggregate;
6


Laboratory trial mixture must not exceed permeability of 2000 coulombs at 56-days of age (ASTM C 1202) and must not exhibit a crack at or before 14 days in the cracking tendency test (AASHTO PP 34).
2.1.2.2 Class HT Specifications
The CDOT Class H and HT concrete have identical specifications and are used for bare concrete bridge decks that will not receive a waterproofing membrane. The difference between the Class H and HT lies in that Class HT concrete is used as the top layer of the bare bridge deck. The specifications for the CDOT Class HT concrete are summarized below:
56-day compressive strength of 4500 lbs./in.2;
Air content of 5% 8% are required;
W/cm ranging from 0.38 0.42;
An approved water reducing admixture;
Must have a minimum of 50 percent AASHTO M 43 size No. 7 or No. 8 coarse aggregate by weight of total aggregate
Laboratory trial mixture must not exceed permeability of 2000 coulombs at 56-days (ASTM C 1202) and must not exhibit a crack at or before 14 days in the cracking tendency test (AASHTO PP 34).
7


2.2 Cracking in Concrete
2.2.1 Importance of Cracking in Concrete
Concrete is known to be weak in tension. In design, concrete beams are assumed to have zero tensile strength. These tensile stresses are fairly low when compared to those experienced by reinforced bridge decks or beams, which spans are restrained between two or more supporting structures. Individual lanes of bridge decks are sometimes placed while others on the bridge remain open for service. For many reasons, bridge decks experience movement (deflection) during daily traffic and thermal expansion which can contribute to the concrete cracking. The earlier the concrete deck cracks the faster the rate of deterioration and need for repair. As a result, the concrete must be more durable and designed to have characteristics that will be advantageous during early ages and in this environment. A decrease in early age cracking will delay the development of corrosion on the reinforcing steel, decreasing its permeability and increasing the structures durability.
Cracking in reinforced concrete structures allows water and contaminants to migrate inside the structure where it can cause deterioration of the reinforcing steel as well as the surrounding concrete matrix. Water that is able to penetrate through the bridge superstructure can also cause damage to the substructure and affect bridge aesthetics.
8


To minimize the amount of cracking and reduce maintenance costs, Class H and HT concrete mixtures was analyzed in this study to ensure the concrete meets the expectations of the CDOT. Further, additional mixtures were evaluated for their effectiveness to eliminate or at least reduce cracking in concrete structures.
2.3 Causes of Cracking in Concrete
2.3.1 Internal Stresses
Concrete cracks as the result of any number of factors. Internal stresses within the concrete are the primary cause of early-age cracking. Internal stresses develop depending upon the heat of hydration, the rate of strength gain, ultimate 28-day and 56-day compressive strength, cement content, percent replacement of cement with supplementary cementitious materials (SCMs), and w/cm (Equation 1);
, water (lbs.)
wl cm =----------------=---------------
cementitious materials (lbs.)
Equation 1
Additionally, the use of chemical admixtures is necessary to create various desirable characteristics of the mixture. These characteristics include reducing shrinkage, delayed set time and air content. All of which can impact the magnitude and rate of development of internal stresses and cause cracking.
9


2.3.2 External Stresses and Normal Use Degradation
Daily, cyclical service loading is a major cause of cracking in concrete bridge decks. These stresses are unavoidable as the Colorado weather, temperature fluctuation, and traveling vehicles gradually degrade the roadways and deck surfaces.
2.3.3 Restraint
Restraint has long been an issue regarding bridge deck cracking. Deck slabs are restrained against movement at joints and internally around steel reinforcement. As concrete expands thermally or shrinkage occurs, the restraint against movement will result in cracking. Expansion joints in bridges help to alleviate cracking due to these stresses.
2.3.4 Shrinkage Strain
Shrinkage strain is a major cause of early age cracking in concrete and the primary focus for this research. Multiple types of shrinkage exist and are all detrimental to the life of the concrete. As water leaves the cement paste matrix, the volume of cement paste begins to shrink and is termed shrinkage.'
10


Drying shrinkage represents the strain caused by the loss of water from concrete. This type of shrinkage results in surface cracking (map-cracking) and causes the surface of the bridge deck to deteriorate at a much faster rate.
A type of drying shrinkage is termed autogenous shrinkage, which occurs as the internal water is gradually depleted during the continued hydration of cement particles over the life of the concrete. Regardless of the type of shrinkage, the volume of the cement paste has a tendency to shrink as the water dissipates. Shrinkage begins to occur immediately after the concrete sets, as surface water begins to evaporate and with the continued hydration of cement particles. The voids in the concrete once occupied by water are then left empty. The volume shrinkage that attempts to occur within the rigid cement paste matrix creates internal stresses within the concrete. These stresses induce a strain on the concrete that results in early age cracking. This research utilizes the AASHTO P34 Restrained Ring Shrinkage Test to measure these shrinkage strains versus time.
The primary objective of this research is to design, batch, and test a minimum of ten concrete mixtures to examine various aspects of concrete mixtures and their influence on cracking. Specifically, this research aims to develop a concrete mixture that is more resistant to cracking than the current Class H and HT specification. A more detailed explanation and understanding of all the tests
11


performed for this research is included in the literature review in Chapter 3 of this thesis.
2.4 Research Objectives
2.4.1 Objectives of Investigation
The primary objectives of this study are to design a more crack resistant concrete for use in Colorados bridge decks. Upon completion, recommendations will be provided the Colorado Department of Transportation for a new low cracking concrete specification or to augment current Class H and HT concrete specifications.
12


Chapter 3
Literature Review
3.1 Preface
This literature review does not examine the effects of super-structure design on concrete bridge deck cracking. Construction practices such as curing, finishing, time of placement (ambient temperature), and consolidation play a major role in bridge deck cracking. This thesis investigates the effect of mixture design factors which influence bridge deck cracking. Curing practices are discussed herein only to emphasize its importance in the practice of placing concrete.
3.2 Curing
Curing is not the focus of this thesis; however, curing is essential to producing quality concrete. Curing is the method used to reduce the evaporation of water immediately after placement and is required to promote continued hydration of the cement, thereby increasing the concretes compressive strength and overall durability. The effect of curing cannot be neglected in practice. Furthermore, the effect of curing on compressive strength and shrinkage cannot be disregarded. All of the research examined for this literature review discusses the importance of
13


adequate curing. Internal curing is the only method of curing pertaining to the scope of this research and is discussed in further detail in Section 3.4.7.3.3
3.3 Concrete Shrinkage
Shrinkage is a major cause of cracking in concrete bridge decks. When cement is hydrated and water evaporates, internal stresses develop and volume shrinkage of the concrete occurs, autogenous shrinkage and drying shrinkage, respectively. The hardened concrete attempts to resist these stresses and cracks as a result. A concrete mixture design may combat shrinkage by adjusting the quantity of any one or multiple materials used in making concrete. A literature review was conducted on several available studies involving cracking in concrete bridge decks. The research information reviewed was built upon in an effort to efficiently provide the CDOT with revised and more durable bridge deck mixture designs.
3.3.1 Effect of Restraint on Shrinkage
Restraint has long been known to cause bridge deck cracking. As a concrete bridge deck dries and moisture evaporates, it experiences a volume decrease termed shrinkage. According to Krauss and Rogalla (1996), the amount of shrinkage depends primarily on the paste content and water content. Reinforcement and the bridge superstructure components such as girders provide restraint against
14


shrinkage, resulting in tensile stresses that cause the concrete to crack (1996). Restrained ring shrinkage tests (AASHTO PP34, ASTM C 1581) allow researchers to conduct a relative comparison of the micro strain associated with different mixture materials at the point of cracking due to restraint in a controlled environment. Cracking is indicated as the point when the strain in the steel ring suddenly decreases. The exposed surface of the concrete ring makes inspection for cracks simple although several mixtures did not exhibit visible surface cracking after the drop in micro strain occurred. The standard specifically states this test is not accurately applicable to field practice or exposed structures. The restrained ring test is not applicable to expansive cements or concrete having a nominal maximum aggregate size (NMAS) greater than 13 mm (0.50 in.). If any of the concrete rings do not crack during the test period, the rate of tensile strength stress development at the time the test is terminated provides a basis for comparison of the materials (ASTM C 1581).
3.3.2 Effect of Curing on Shrinkage
Although multiple methods of curing are not included in the scope of this research, the method used to cure concrete is essential to its characteristics such as durability, rate of strength gain, ultimate strength, freeze/thaw resistance, and appearance. Cement paste will never completely hydrate when the w/c ratio is below 0.42. A
15


layer of C-S-H builds up on the largest grains of cement and hinders the hydration process. Curing helps ensure as much hydration as possible occurs and at a reasonable cost (Mindess, Young, and Darwin, 2003). After meeting with the CDOT, it was discovered that training on the importance of curing techniques was non-existent, leaving a huge opportunity for project error. A survey of other state DOTs further strengthened the widespread belief suggesting curing practices are a major cause, perhaps the primary cause, of transverse deck cracking. Krauss and Rogalla performed their own survey of existing DOTs fifteen years ago (1993) and received many of the same responses concerning curing. They discovered many curing practices were being used in different states depending upon the job but that no standard curing practice existed for bridge decks. Practices ranged from allowing only membrane or curing compounds to requiring long-term wet curing using curing compounds, and in many cases, the contractor was given the liberty to choose the method. Krauss and Rogalla suggest the latter practice will most likely result in problems with the concrete. Typically, the contractor would choose the cheapest method to save money, but the cheapest method is not typically the most effective one for the job. Babaei and Hawkins (1987) point out that fogging or evaporation retarding films substantially reduce early plastic deck cracking if applied immediately after strike-off of the concrete. In addition, Babaei and Hawkins suggest applying wet burlap as soon as possible. This method results in
16


fewer smaller cracks than curing compounds; delayed water curing even increases cracking.
Krauss and Rogalla (1993) reported high cement content concrete to be most affected by curing. Concrete with a w/c ratio equal to 0.50 and cement
j >
content of 278 kg/m (470 lb/yd ) that was wet cured for 60 days experienced little
change in time to first cracking of the ring in the restrained ring shrinkage test.
When the w/c ratio was lowered to 0.35, cement content increased to 501 kg/m3 >
(846 lb/yd ), and curing remained the same, time to first cracking of the ring increased from 11.7 to 21.0 days.
Mindess, Young, and Darwin (2003) suggest the duration of and the maximum temperature reached by the cement paste plays a major role in cracking. They report pastes which achieve elevated temperatures during curing experience reduced irreversible shrinkage with no effect on reversible shrinkage. A paste exposed to 65C (150F) reduces irreversible shrinkage by 66.67% and total shrinkage by 33.33%. This reduction is attributed to the large proportion of the capillary porosity having formed as macro pores, resulting in a reduced micro porosity of C-S-H. The effective reduction in shrinkage is a function of the duration of exposure time to higher temperatures. According to Mindess, Young, and Darwin the exposure time necessary to reduce shrinkage can be relatively short and is often less than the total curing time (2003).
17


Wet curing techniques such as quickly applying wet burlap, water ponding, or continuous water misting are all beneficial curing methods that reduce cracking by reducing the evaporation rate of water in concrete. High performance and high cement content concrete only have a small amount of mixture water to evaporate. Wet curing not only slows down the rate of water evaporation but cools the concrete simultaneously. This results in lower thermal stresses that develop due to the heat of hydration (Krauss and Rogalla, 1993).
Mixed opinions exist as to what is the ideal curing method. Krauss and Rogalla suggest the immediate use of windbreaks and wet curing the concrete. Curing should consist of misting, curing compound, and wet burlap. The minimum curing period is 7 days, ideally 14 days, when the evaporation rate exceeds 1 kg/m2/hr (0.2 lb/ft2/hr) for normal concrete and 0.5 kg/m2/hr (0.1 lb/ft2/hr) for concrete susceptible to early-age cracking due to low w/c ratios. They report that exposure to high temperatures after the curing period is complete can also help to reduce irreversible shrinkage. Most researchers agree that a standardized method of curing is needed and should be instated by AASHTO.
Deshpande et al (2007) examined the effect of the curing length on air-entrained concrete made with both Type I/II and Type II coarse ground cement. Concrete made with Type I/II cement exhibited significantly increased shrinkage when comparing curing durations at different periods of time beyond initial drying.
18


At 30 days beyond initial drying the shrinkage of concrete cured for 3, 7, 14, and 28-days were 500gs (micro strain), 375, 340, and 274g£, respectively. As the curing period increased, the free shrinkage decreased. This trend continued through measurements taken up to 365 days past initial drying. At 365 days past drying the largest difference in shrinkage strain occurred between concrete cured for 3 and 7 days, 690 and 515jas, respectively. Differences in strain were small between concrete cured for 7 and 14 days at 525 and 500pe, respectively. Air-entrained concrete made with Type II coarse ground cement exhibited a similar trend; shrinkage decreased with increased curing periods. At 30 days past drying, concrete cured for 3, 7, 14, and 28-days experienced free shrinkage micro strains of 250, 205, 110, and 5g£, respectively. Concrete cured for 3 days experienced slightly more shrinkage than concrete cured for 7 days until approximately 75 days past drying. After that point the difference in free shrinkage results were relatively small. At 180 days past drying a difference of approximately 50p£ existed between the concrete cured for 3 to 7 days and those cured for 14 to 28-days. It is apparent from the results that an extended curing period creates a more durable concrete for both Type I/1I and Type II coarse ground cement concrete. It is clear that the ultimate shrinkage of concrete made with Type I/II cement is significantly higher than Type II coarse ground cement concrete at all ages. Free shrinkage measurements were taken at intervals of 30, 180, and 365 days past drying on
19


concrete cured for 3 days, and a difference of 225, 240, and 300pc, respectively, existed between the Type I/II and Type II coarse ground cement. The research performed by Deshpande et el (2007) clearly shows the advantage of using Type II coarse ground cement over a Type I/II cement when the effect of curing periods on shrinkage are being considered.
3.3.2.1 Internal Curing Using Light Weight Aggregate
The use of new presoaked lightweight aggregate (LWA) in high performance concrete (HCC) is becoming more common. The aggregate is said to internally cure as a result of it being soaked before batching and contributes to the hydration process instead of absorbing water from the concrete mixture. This approach uses aggregate made of porous expanded shale, sufficient to provide effective internal curing in order to reduce self-desiccation and autogenous shrinkage cracking. Cusson and Hoogeveen conducted research (2006) at the Canadian Institute for Research and Construction examining high performance concrete made with Type I Portland cement and replacements of sand by the light weight aggregate. A control mixture was designed with a cement-sand-stone ratio of 1:2:2 and w/cm equal to 0.34. It is noted that the water used to pre-soak the LWA was accounted for in the calculation of the w/cm and remained constant for all of the concretes examined. This requirement was said to have made the evaluation of the internal curing
20


effectiveness more severe than if additional water had been used to soak the aggregate. The three batches substituted normal weight sand with 6, 12, and 20% pre-soaked LWA and a fourth control mixture substituting 0% LWA. One large concrete prism 200 x 200 x 1000 mm (8 x 8 x 40 in.) was cast for each mixture with reinforcement and used a setup attaching strain gauges to the steel in order to determine the restrained shrinkage. A second concrete prism of the same size was cast from each mixture without reinforcement and used for unrestrained shrinkage testing. This prism was cast with thermal couples and relative humidity (RH) sensors (measuring self-desiccation) implanted within the fresh concrete. Compressive strength and splitting tensile strength tests were also performed on 100 x 200 mm (4x8 in.) cylinders. The 20% LWA concrete experienced reduced drying due to the internal curing. The RH of the control specimen reduced from 100% at set time to 98% after 2 days and 96% after 7 days. The RH of the 20% LWA concrete reduced to 98% after 2 days and 94% at 7 days. The control test specimen had a 7 day compressive strength of 50MPa (7252 lbs/in ) versus the 20% LWA concrete of 57MPa (8267 lbs/in2). Cusson and Hoogeveen attribute this to the improved hydration of the pre-soaked LWA. Free shrinkage test results prove that as the LWA content increased in the concrete mixtures the autogenous shrinkage decreased. The 0, 6, 12, and 20% LWA concretes experienced strains of 252p.s (micro strain), 210, 112, and 46pe respectively at 2 days of age. After
21


restrained shrinkage tests were performed the stress/strength curves were normalized. This was done so the comparison could be made between the various curves corresponding to different concretes, which require different degrees of restraint during testing. Restraints varied from a low 0.9% for the 0% and 6% LWA concrete in order to avoid failure, to a high restraint of 1.1 for the 20% LWA concrete, having the loading system slightly pulling on the prism. The replacement of sand with LWA increased the modulus of elasticity (MOE) considerably. At 3 to 4 days of age, the MOE was several thousand MPa higher for the 20% LWA concrete than the control (0% LWA) concrete. The 7 day splitting tensile strengths were measured to be 4.1 MPa (595 lbs/in2), 4.8MPa (696 lbs/in2), 4.5MPa (653 lbs/in2), and 4.2MPa (609 lbs/in2) for the 0%, 6%, 12% and 20% LWA concretes respectively. The maximum stress/strength ratio achieved by the 20% LWA concrete was 50% after nearly 3 days. These results illustrate the LWA to be extremely beneficial in reducing cracking. Cusson and Hoogeveens research shows how effective internal curing is against shrinkage and tensile stress in concrete, especially high performance concrete. Their results prove the effect of LWA sand replacement on strain and stress reductions. Their data indicates that a 25% LWA concrete could possibly eliminate autogenous shrinkage and tensile stress. Significant swelling did occur in the 20% LWA concrete. As a result, it is
22


not recommended to use more than a 25% LWA concrete because of the possibility of excess swelling (Cusson and Hoogeveen, 2006).
3.4 Design Mixture Factors Affecting Cracking in Concrete
3.4.1 Silica Fume
Substitution of cement with silica fume produces a denser concrete matrix. It results in a more rapid rate of hydration, which is accompanied by a higher heat of hydration and increased early strength development (Transportation Research Circular E-C107, 2006). A higher heat of hydration results in higher thermal stresses and reduced bleeding, making concrete more prone to plastic shrinkage (Xi et al, 2003). Another study by Bissonnette, Pierre, and Pigeon (1999) also claims silica fume is not beneficial in concrete for reducing cracking. One of their research programs compared two concrete mixtures with w/cm equal to 0.33. One of the mixtures contained 15% silica fume substitution for portland cement. Restrained ring shrinkage tests were performed and the silica fume concrete produced an additional 300 micro strains at 4 days over the 100% portland cement concrete. Bissonnette et al concluded that the presence of silica fume in concrete results in an increase in long term shrinkage. However, the resulting early age increase in shrinkage leads to significant cracking because the tensile strength is so low at early ages (1999).
23


Whiting, Detweiler, and Lagergren (2000) also researched the effect of silica fume on concrete shrinkage in full depth decks and concrete overlays. Full depth mixtures used lower cementitious material contents and air contents with higher w/cm than the overlay design mixtures. Silica fume substitution ranged from 0 to 12 percent of the total cementitious material weight and w/cm for overlays ranged from 0.30 to 0.35; full-depth decks w/cm ranged from 0.35 to 0.45. Unrestrained drying shrinkage tests AASHTO T 160 (ASTM 157) were performed on three 75 x 75 x 285 mm (3 x 3 x 11.25 in.) prisms molded for each mixture. The unrestrained test specimens were cured in lime saturated water; full-depth mixtures were cured for 7 days and overlay mixtures only 3 days. They were then moved to a controlled relative humidity of 50% and a temperature of 23 C (73 F).
Restrained shrinkage tests were performed per ASTM C 1581 (AASHTO PP34) on a 75 mm (3 in.) thick, 150 mm (6 in.) high concrete ring around the outside of a 19 mm (0.75 in.) thick steel cylinder having an outside diameter of 300 mm (11.75 in.). The restrained ring specimens were cured for periods of 1 and 7 days, intending to represent both the worst and best field curing practices. A data acquisition system wired to four strain gauges that were attached (90 offset) around the inside of the steel ring measured the strains at thirty minute time increments. Their results show the presence of silica fume to have little effect on long term shrinkage (450 days). Early age shrinkage (4 days) was higher for
24


concrete mixtures with silica fume, versus the control mixtures made without. At this age, results consistently show an increase in shrinkage with increased silica fume content. Lower w/cm concrete mixtures (0.36) demonstrated less shrinkage when made with a constant replacement of silica fume (1.8%) than concrete with a higher w/cm (0.43). The lower the w/cm the smaller the cement content relative to the mixture. The smaller the amount of cement in the mixture means the smaller the amount of cement paste formed that is available to shrink. Whiting et al point out that the two mixtures having w/cm of 0.36 and 0.43 had paste volumes of 25.2 and 27.5%, respectively. It is also noted that small variations in w/cm may greatly influence shrinkage in concrete. The silica fume specimens cured for one day cracked sooner than control specimens. For specimens cured 7 days, the silica fume in the concrete significantly reduced time to first crack. Whiting et al suggest not exceeding 6% silica fume replacement of portland cement because it begins to have an adverse negative effect on shrinkage and cracking.
3.4.2 Fly Ash
Research concerning the replacement of portland cement with fly ash in a concrete mixture has returned contradicting results. Class F and class C fly ash replacement is a very effective method of slowing the rate of C-S-H growth. It reduces early age strength gain and early concrete temperatures and still achieves the same
25


ultimate strength (Xi et al, 2003). High volumes of fly ash substitution for portland cement have been studied in the past. Atis reported a decrease in drying shrinkage with the use of fly ash (2003). They created mixtures with varying w/cm (0.28 to 0.34) which they had previously determined to be optimal for maximum compact-ability using the vibrating slump test (Cabrera and Atis, 1999). These optimal w/cm were used in creating zero slump concrete mixtures and achieving workability by using a carboxylic type water super-plasticizer. The mixtures were designed containing 100% (control mixture), 50%, and 30% portland cement replacement with a low calcium class F fly ash (ASTM C 618). Two molds of each mixture were made to be tested. The mixtures in the molds were the same except one used a super-plasticizer. They performed unrestrained shrinkage tests on 50 x 50 x 200 mm (2x2x8 in.) concrete prisms that had been unmolded after 24 hours and then stored at 20 C (68 F) and a relative humidity of 65 percent. Measurements were taken up to six months of age to determine changes in length (drying shrinkage) using a mechanical dial gage. The super-plasticized mixtures containing 0, 50, and 70% fly ash replacement of portland cement exhibited strains equal to 385, 263, and 294 micro strain respectively (2003). When these were compared with the same percent fly ash replacement mixtures but not containing a super-plasticizer, they exhibited approximately 50% less shrinkage. The compressive strengths were measured and compared between the control mixture
26


and the fly ash concrete. The compressive strength of 50% fly ash concrete exceeded the control concrete once it reached 7 days of age. The compressive strength of the 70% fly ash concrete was exceeded by the control at all ages. The 28-day compressive strengths showed a drastic difference. The control, 50% fly ash concrete, and 70% fly ash concrete compressive strengths were 65MPa (9430 psi), 67MPa (9720 psi), and 31 MPa (4500 psi) respectively. Cabrera and Atis research suggests concrete mixtures with portland cement replacement by approximately 50 percent fly ash and no super-plasticizer to be optimum.
Research conducted at the Materials Laboratory at CU-Boulder has shown concrete made with smaller particles of fly ash certainly have some advantages over conventional concrete, but may not be applicable for bridge decks due to its high early strength, high ultimate strength, and low crack resistance (Xi et al, 2003). Some studies say both Class C and Class F fly ash replacement in concrete increases drying shrinkage and results in increased early cracking with decreased development of tensile strength (Hadidi and Saadeghvaziri, 2005). The research studied in the literature review is tough to decipher; fly ash replacement have mixed results. Its reduction in the rate of stiffness development is helpful in reducing its potential for cracking (Transportation research circular E-C107, 2006). While the reports are contradictory, the majority of the literature suggests fly ash is beneficial with regards to concrete shrinkage.
27


3.4.3 Water to Cementitious Materials Ratio (w/cm)
The w/cm is the ratio of the weight of the water to the weight of all cementitious materials per cubic yard of concrete (Equation 1). This ratio effects concrete in many ways. The permeability, porosity, ultimate strength and rate of strength gain are all affected by changes in the w/cm. It is generally accepted that drying shrinkage increases significantly as the water content increases. ACI 224 Report
'j i
states that a typical concrete specimen, 134 kg/m (225 lbs/yd ) water content, resulted in a drying shrinkage of approximately 300 micro strains. It also states that drying shrinkage increases at a rate of 30 micro strain per 5.9 kg/m3 (10 lbs/yd3) increase in water content. A study of 12 bridges in Pennsylvania reported crack intensities of 0 to 87m/100m (265 ft/1000ft) with mixture water contents varying from 158 to 173 kg/m3 (267 to 292 lbs/yd3). An increase in water content showed increased drying shrinkage of approximately 75 micro-strains, indicating that with respect to transverse cracking, mix water content alone was not the significant difference in the performance of bridge decks (Babaei and Purvis, 1995a). Similar articles report concrete with a w/cm greater than 0.45 tend to have high porosity and can exhibit substantial drying shrinkage, which results in reduced protection of the reinforcing steel from chlorides (Transportation research circular E-C107, 2006).
28


3.4.4 Cement Content
The cement content has a significant effect on shrinkage and cracking in concrete. Concrete made with higher cement content and a low w/cm is more susceptible to cracking than concrete with low cement content and higher w/cm (Xi et al, 2003). Xi et al research and other literature suggest limiting cement content to 470 lbs/yd (279 kg/m3) and that a cement paste volume less than 27.5 % can significantly reduce cracking. However, as high strength concrete has become more common in the industry, it is often encouraged to increase the cement content. Proper measures must be taken for concrete made with increased cement content or it can significantly increase cracking (Transportation research circular E-C107, 2006).
Deshpande et al (2007) conducted research using Type II coarse ground Portland cement in nine concrete mixtures while varying w/cm and aggregate content. It was concluded that a clear trend for shrinkage results from variations in w/cm ratio did not exist. At 180 days of age a seemingly sensible pattern of shrinkage occurred in the concrete having the highest aggregate content (80%).
The higher the w/c the more shrinkage that occurred; 280pe: w/c = 0.40 and 305pi;: w/c = 0.50. This wasnt the case for the mixtures containing lower aggregate contents of 60 and 70%. They reported the greatest shrinkage occurred in the concrete with a w/c equal to 0.40 (the lowest w/c) and a 60% (lowest) aggregate content. Shrinkage was lowest in the concrete with a w/c equal to 0.40 and having
29


the highest aggregate content of 80%. The research is consistent with other literature in stating that for a given w/c ratio, the shrinkage decreases with an increase in aggregate content. The aggregate acts to restrain the concrete against shrinkage. Adversely, for a given aggregate content, the results of this study show changes in the w/cm and using coarse ground cement to have very little affect on shrinkage (Deshpande et al, 2007).
3.4.5 Cement Type
3.4.5.1 General Effects of Cement Fineness
Cement types vary depending upon the use or the project. Different types of cement produce different temperatures as a result of their hydration processes. Some cement is ground finer and others are coarser. Some cement is designed for high early strength, resulting in a high heat of hydration and high thermal stresses. The resulting stresses make concrete more likely to crack. There are also types of cement designed to gain strength more slowly, corresponding to a lower heat of hydration (Type I/II, Type II, and Type IV). Concrete made with these types of cement is expected to result in lower thermal tensile stresses and reduced cracking. Burrows (2003) reports that cracking in bridge decks increased in 1973 when the building code increased 28-day compressive strength requirements from 3000 lbs./in.2 to 4500 lbs./in.2. The increase in the rate of strength gain causes concrete
30


to become more brittle and likely to crack. Burrows points out that in 1966
'y
Virginia increased its 28-day compressive strength requirements from 3000 lbs./in. to 4000 lbs./in.". It was at this time when bridge deck cracking increased from 11% to 29%. His research brings attention to numerous Colorado area bridge decks built in the 1950s that are still in great condition but are being demolished to accommodate a necessary widening of Interstate-25. The bridges of the 1950s had unacceptable 28-day compressive strengths by todays code, but have significantly maintained their structural integrity for half a century. As of 1995, Burrows reports bridge deck cracking in the United States to have increased to 52% of all bridges. This clearly illustrates the upward trend of bridge deck cracking as the required strengths and rate of strength gain continue to increase (Burrows, 2003).
Xi et al suggest using Type II cement and avoiding finely ground cement and/or Type III cement (2003). Cements with high alkali content, high C3S and C3A contents, low C4AF, and high fineness have an increased development of strength and are therefore more likely to crack. This is another reason the research circular raises caution in using Type III cement for bridge decks (Transportation research circular E-C107, 2006).
31


3.4.5.2 Coarse-Ground Cement
Research conducted by Deshpande et al for the University of Kansas Research Center show significantly reduced shrinkage in concrete using Type II coarse-ground cement versus Type I/II cement. In addition, Deshpande examined the effect of aggregate content and w/c on shrinkage (2007). A program consisting of three concrete mixtures made with Type I/II portland cement and three mixtures made with Type II coarse-ground cement. The w/c was 0.40 for all six mixtures but the aggregate content varied between 60, 70, and 80% for each type of portland cement. At 180 days of age, free shrinkage tests measured significantly lower shrinkage strains (280pe) in the Type II coarse-ground cement having the highest aggregate content (80%) than the shrinkage strain (665pe) measured in the concrete made with the Type I/II portland cement having the lowest aggregate content (60%). These strains tapered off near 180 days of age and, at 365 days of age, illustrated an insignificant amount of continued shrinkage strain (Deshpande et al, 2007). This suggests that both Type II coarse ground cement and mixtures with a higher aggregate content are more suitable for use in crack resistant concrete bridge decks.
Brewer and Burrows (1951) tested three cement clinkers ground to finenesses ranging from 1200 to 2700 cm.2 (186 to 419 in.2), in 300 cm.2 /g increments. They performed tests similar to ASTM C 1581 but using an apparatus created before the
32


standard was adopted by ASTM. They also performed unrestrained shrinkage tests on mortar bars. Restrained shrinkage tests showed concrete made with coarse-ground cement resisted cracking longer than the more finely-ground cement concrete. When the coarse-ground cement rings cracked, the unrestrained shrinkage mortar bars were examined further. They discovered the coarse-ground cement mortar bars shrank forty percent more and at a slower rate than the more finely-ground cement concrete. In the end the coarse-ground cement concrete shrank as much as forty percent more than the more finely-ground cement bars due to drying shrinkage. It was concluded that mortars made with coarse ground cement are significantly more resistant to cracking than more finely-ground cement concrete due to drying shrinkage (Brewer and Burrows, 1951).
3.4.5.3 Shrinkage Compensating Cements
Shrinkage compensating cements (SCC) are another cement type currently being studied in the United States. Type K cement (ASTM C845-80) creates an amount of expansion when the concrete is hardening, in an effort to counteract autogenous shrinkage and drying shrinkage. ACI 223R-90 (1992) illustrates the specifications concerning the use of expansive cement. According to Xi et al, the problem with designing concrete using expansive cement is predicting the amount of expansion necessary for each individual project (2003). Krauss and Rogalla performed
33


research using Type K cement and reported two specimens used in restrained ring shrinkage testing didnt experience significant cracking (1996). Surface cracking occurred but no distinct cracks developed. The research shows ring strains decreased to a constant level without cracking. They also examined a SCC containing an ettringite forming additive. Control mixtures cracked at an average of 20 days and the SCC concrete time to cracking extended to around 36 days. Researchers state the restrained ring shrinkage test has merit but field performance will vary from laboratory results.
Perragaux and Brewster investigated several bridge decks for the New York State Department of Transportation in 1992. Results varied as they compared the bridge decks made using shrinkage compensating cement with surrounding structures previously made using Type II cement. In some structures it was believed that SCC reduced shrinkage by 25% and in some cases the SCC structures cracked more than the Type II cement structures. The research concluded shrinkage compensating cement is not advantageous when compared to Type II cement (Perragaux and Brewster, 1992).
Studies performed by the Ohio and New York State Departments of Transportation have returned mixed reviews concerning shrinkage compensating cement. Ohio reported success with the cement in bridge decks while New York had issues with durability (Philips et al, 1997). In 1989, Purvis performed research
34


on concrete slabs made with SCC and found the final net drying shrinkage of SCC slabs was less than slabs made with Type I cement, but the SCC slabs experienced more creep.
3.4.6 Aggregate Content
Because shrinkage is mostly a paste property, it makes sense that increasing the aggregate content decreases shrinkage. Aggregates help by providing restraint to shrinkage while occupying space within the concrete matrix; a space that would otherwise be occupied by additional cement paste. This also helps to create a more economical project because cement is the most expensive material used in making concrete (Transportation research circular E-C107, 2006). However, aggregates themselves may be responsible for shrinkage. The use of highly absorptive aggregates has proven to result in increased shrinkage. They are more compressible and therefore allow for higher shrinkage. Some may shrink an appreciable amount themselves by the time they are completely dry (Transportation research circular E-C107, 2006).
Studies performed by Deshpande et al (2007) examined a program consisting of nine concrete mixtures made with Type II, coarse-ground portland cement. The w/c were 0.40, 0.45, and 0.50 and the aggregate contents were 60, 70,
35


and 80%. Three mixtures were made with a 0.4 w/c and varied the aggregate content from 60, 70 and 80% and this was done for all three w/c.
It is clear that a delicate balance of aggregate content and w/c ratio are necessary to determine the most appropriate concrete design mixture for crack resistant bridge decks. At 180 days of age, a trend developed showing significantly less shrinkage occurring in the two concretes made with the higher aggregate content (80%) and the lower aggregate content (60%) concrete. The 80% aggregate concrete experienced less shrinkage than the 60% aggregate content concrete. The smallest strain (280ps) was produced by the highest aggregate content mixture having the lowest w/c. Accordingly, the highest strain (305jus) produced in the three mixtures was in the concrete made with an aggregate content of 80% and a w/c of 0.50 (highest w/c). This was not the case with the other six mixtures. The mixtures containing an aggregate content of 60% and 70% did not follow the same trend as the concrete made with an aggregate content equal to 80%. Although the three mixtures with a 70% aggregate content produced significantly lower strain (360-380pa) than the 60% aggregate content mixtures (450-51 Ops), it wasn't the mixture with the highest w/c producing the largest amount of shrinkage (Deshpande et al, 2007).
36


3.4.7 Aggregate Composition
3.4.7.1 General
The type of aggregate used in concrete can also affect shrinkage. Tests have shown quartzite aggregate to exhibit significantly lower shrinkage strains than concrete made using limestone aggregate. When comparing concrete made using granite, limestone, and quartzite aggregate, the shrinkage strain values at 30 days were 283, 320, and 340 micro strain respectively. These results show that granite aggregate allows for the least amount of shrinkage (Deshpande, Darwin, and Browning,
2007).
Xi et al also state that the larger the maximum size of the aggregate the smaller the resulting shrinkage. They report that when the cement paste shrinks, it cannot pull the larger surrounding aggregate closer since they are already in close contact. Micro cracks will develop but so long as they dont develop into larger cracks, the concretes ability to resist cracking is considered to be enhanced and shrinkage reduced (2003).
3.4.7.2 Aggregate Composition and Water to Cementitious Materials Content
Similar research conducted by Meyerson, Mokarem, and Weyers for the Virginia Department of Transportation (2003) used three types of aggregate; limestone, gravel, and diabase. Type I/II portland cement mixtures with no SCMs were
37


examined in the first three programs. Each program consisted of a range of w/c. The fourth program of mixtures were designed with cement replacements by 40% (by weight) ground granulated blast furnace slag (grade 120), Class F fly ash, and a pure amphorous micro-silica, each conforming to their appropriate standards ASTM C 989-98, ASTM C 618-97, and ASTM C 1240-97, respectively. At 7, 28, and 90 days of age compressive strength tests were performed following ASTM C 39-98 and 102 mm x 203 mm (4x8 inch) test cylinder specimens were fabricated according to ASTM C 192-98. A trend of compressive strength developed with 100% portland cement mixtures having varying w/c, ranging from 0.42 to 0.49.
The first program had w/c of 0.49, 0.47 and 0.46 and the second program consisted of w/c of 0.45, 0.43, and 0.42 and both incorporated limestone, diabase, and gravel mixtures respectively. In Mokarem et als research (2003) the gravel aggregate concrete mixtures consistently had the highest compressive strength, but in most cases it was not significantly stronger than the diabase aggregate. However, the limestone aggregate mixtures consistently produced significantly lower compressive strengths than both the diabase and gravel mixtures. As expected, the compressive strengths correlated to the w/c, the highest compressive strength correlating to the concrete having the lowest w/c and the lowest compressive strength correlating to the concrete having the highest w/c. These results were then tested with a third program of 100% portland cement mixtures to verify that it was
38


not the limestone aggregate properties alone that caused a reduced compressive strength. The program included mixtures having w/c ratios of 0.33, 0.35, and 0.39 for the limestone, gravel and diabase mixtures, respectively. These are the lowest w/c of any of the programs and this is the largest variation of w/c ratios examined in Mokarem et als research (2003). At 7, 28, and 90 days of age the limestone mixtures compressive strength significantly exceeded that of the gravel and diabase mixtures. At 7 days of age, the compressive strengths measured 7150, 6260, and 6070 lbs/in.2 (503, 440, and 427kg/cm.2) for the limestone, gravel, and diabase mixtures, respectively. At 28 and 90 days of age the compressive strength continued to follow this trend. These results illustrate the inverse proportionality between compressive strength and w/c.
3.5 Unrestrained Shrinkage Test
Mokarem et al later performed standard tests to determine length change according to standard ASTM C 157-98. Recall, the first program of mixtures had w/c of 0.49, 0.47, and 0.46 for limestone, diabase, and gravel mixtures, respectively. The diabase aggregate concrete experienced the greatest percent length change at almost every age, although the percent changes in length between the three aggregate type mixtures were insignificant up to 56-days of age. The following unrestrained shrinkage data references programs one, two and three and the limestone, gravel,
39


and diabase concrete mixtures, respectively. At 56-days of age, program one percent length changes were -0.0380, -0.0367, and -0.0392, program two percent length changes were -0.0342, -0.0323, and -0.0392, and program three percent length changes were -0.0321, -0.0328, and -0.0364. After 56-days of age, the diabase aggregate mixtures made with only portland cement began to experience significantly greater percent changes in length than both the limestone and gravel aggregate mixtures, while they continued to experience insignificantly different percent changes in length from one another. The results clearly show an increase in rate of length changes at later ages. At 120 days of age, program one percent length changes were -0.0431, -0.0432, and -0.0490, program two percent length changes were -0.0401, -0.0384, and -0.0457, and program three percent length changes were -0.0367, -0.0380, and -0.0453. At 180 days of age, program one percent length changes were -0.0468, -0.0462, and -0.0541, program two percent length changes were -0.0442, -0.0419, and -0.0514, and program three percent length changes were -0.0394, -0.0415, ad -0.0494. Recall the second program consisted of mixtures with w/c of 0.45, 0.43, and 0.42 (the middle range of program w/c examined) for the limestone, diabase, and gravel mixtures, respectively. When standard changes in length were measured for this program, the diabase again experienced the greatest percent length change. These tests show something interesting. The gravel and limestone mixtures experienced percent length changes
40


correlating to their w/c ratio. The limestone concrete having a w/c of 0.45 experienced a greater percent length change than the gravel concrete having a w/c of 0.42. These results support the idea that a higher w/c equates to more water in the mixture and therefore, more shrinkage. However, the diabase concrete had a w/c (0.43) in the middle of the three mixtures and yet it experienced a significantly larger percent length change. When the third program having the lowest range of w/c was examined, the unrestrained shrinkage results illustrate a trend correlating the highest w/c (diabase, 0.39) to the largest percent length change, and the lowest w/c (limestone, 0.33) to the smallest percent length change. Mokarem et al attribute this to the to the diabase aggregate absorption value of 1.04%, versus the limestone and gravel aggregate which had absorption values of 0.48% and 0.75% respectively. These values indicate that the diabase has more voids filled with water than the other aggregate, which can increase drying shrinkage (2003).
When comparing the SCM mixtures, researchers examined mixtures containing the same type of diabase aggregate and the same w/c ratio. The mixtures containing fly ash experienced the greatest shrinkage. Micro silica and Ground Granulated Blast Furnace Slag (GGBFS) mixtures were insignificantly different from one another. The drying shrinkage in the mixtures containing SCMs exceeded that of the 100% portland cement mixtures being compared against. This is possibly due to the denser concrete matrix created when using
41


SCMs. Capillary voids are smaller and would exude less water than normal, larger capillary voids according to Mokarem et al. This is where drying shrinkage primarily occurs (Mokarem, et. al., 2003).
3.6 AASHTO PP34 / ASTM C 1581
In the ASTM C 1581 (AASHTO PP34), Standard Test Method for Determining Age at Cracking and Induced Tensile Stress Characteristics of Mortar and Concrete under Restrained Shrinkage, a concrete ring is cast around a steel ring. Before it was adopted as a standard by ASTM, dimensions of both the steel and concrete ring for the test were modified for various reasons. The current standard (AASHTO PP34, ASTM C 1581) specifies the steel ring to have a wall thickness of 0.50 +/-0.05 in. (13 +/- 0.12 mm), an outside diameter of 13.0 +/- 0.12 in (330.0 +/- 3.3 mm), and a height of 6.0 +/- 0.25 in. (152.0 +/- 6.0 mm), machined smooth on all surfaces. The concrete ring molded around the steel ring is 1.50 in. (38.0 mm) thick. The specimens must be transferred to the testing environment within ten minutes of completion of casting. Four strain gauges are mounted at mid-height (offset 90) around the inside of the steel ring. A data logger begins recording strain measurements within two minutes of the rings being placed in the testing environment. As the concrete ring experiences shrinkage (volume decrease), stresses develop resulting from the steel ring restraining the concrete. The time and
42


micro strain is recorded upon start and micro strain values are recorded by a data acquisition at intervals not to exceed 30 minutes. Moist curing of the molds must begin within 5 minutes of the first strain reading. Moist curing continues for twenty-four hours using wet burlap, a relative humidity of 50% +/- 4%, and at a temperature of 73 +/- 3.5 F. Micro strain averages are recorded at pre-determined days of age and cracking is recorded to the nearest 0.25 day. When cracks occurs the most recently recorded micro strain prior to cracking is examined. This reading is used as a basis for equations which estimate the micro strain at the actual time of cracking.
Over time, variations of the ring test have been performed. The dimensions of the rings used for the test were altered several times. Krauss and Rogalla (1996) examined the affect of changing the dimensions of the rings used for the test. They placed shrinkage stresses that were both uniform and stress increasing linearly from the interface between the concrete and the steel, on the steel ring. They expected this to represent circumferential surface drying or drying from either the top or bottom surface. The research discovered that the height of the steel rings affected the shrinkage stresses in the concrete. As the height increased from 76 mm (3.0 in.) to 152 mm (6.0 in.), shrinkage stresses were reduced. Krauss and Rogalla varied the ring thickness from 13.0 mm (0.50 in.) to 25 mm (1.0 in.) but found little difference in the shrinkage stresses or cracking tendency. Thinner steel rings were
43


associated with higher stresses in the steel and the stresses in the concrete rings increased as they increased the steel ring thickness (1996).
Attiogbe et al examined ring data involving the thickness of the concrete ring versus its time to cracking. They discovered that the concrete ring thickness was linearly proportional to its time to cracking and that the depth of drying increases proportionally with the square root of drying time (2004). ASTM C 1581 (AASHTO PP34) is regarded by the engineering field to be a valid and extremely valuable standardized test to determine the durability of concrete, especially when considering concrete cracking in bridge decks.
3.6.1 Restrained Ring Shrinkage Test
The restrained ring shrinkage testing (AASHTO PP34-98) was performed for a period of 180 days of age, on 42 ring specimens, and strain measurements recorded at 7, 28, 56, 90, 120, 150, and 180 days of age (Mokarem et al, 2003). Average strains were calculated at each of these days and equations based upon the most recent strain record prior to cracking were used to estimate the strain at any other day. In the first program of mixtures, the diabase rings never cracked through the end of the test period. At the end of 180 days, the diabase concrete rings had an average micro strain value of -132ps, significantly less than the limestone and gravel concrete rings. The limestone and gravel concrete rings cracked at 125 days
44


and 117 days, respectively. At cracking, the limestone ring was determined to have an average micro strain value of -234pe at approximately 120 days, meaning it cracked when it reached a value slightly higher than -234pe. When the gravel concrete ring cracked it had an estimated micro strain value of -21 Ops. Mokarem et al report the diabase concrete had a lower modulus of elasticity (MOE) than the limestone and gravel concrete and researchers believe this may have been why the diabase concrete didnt crack. Mokarem et al state that a higher modulus of elasticity concrete is stiffer and possibly able to resist shrinkage in an unrestrained condition, but the stiffer concrete may create higher strains on the ring in a restrained condition. The mixtures from program two didnt crack at all. At 180 days, the average micro strain values for the limestone, gravel, and diabase concrete were -168, -194, and -200ps, respectively. Again, the modulus of elasticity is possibly the cause for the trend in micro strain. The concrete associated lowest MOE having the restrained shrinkage strains. The third program had the lowest of the w/c for the limestone, gravel, and diabase concrete mixtures. Only the gravel and diabase experienced cracking at 165 and 172 days, respectively. The diabase and gravel concrete rings both had an estimated micro strain value of -21 Ope at cracking. Researchers attribute this programs trend in cracking to the w/c. Lower w/c should theoretically experience less shrinkage. In the third program the w/c ratio was the lowest for the limestone, which experienced
45


a significantly lower amount of strain than the diabase and gravel concrete. None of the rings from the fourth program cracked during the 180 day test period. These mixtures contained SCMs and experienced lowest strains of any programs mixtures. Average micro strain values ranged from -142pe to -193pe for all of the mixtures at 180 days of age. Mokarem et al note that the strain measured for the fly ash concrete was the highest for both the restrained and unrestrained shrinkage tests. The slag concrete measured the lowest average micro strain value at the end of the test period. Researchers looked at data for the only four rings that broke and each ring had a micro strain greater than -200ps. Therefore, it was estimated that micro strains greater than -200ps will result in cracking of restrained drying shrinkage rings. Using data obtained from concrete having an average micro strain value of -200pe, it was determined that a strong correlation existed.
3.7 Length Change
The corresponding length change associated with concrete having restrained shrinkage strain measuring -200ps was thought of as a standard. A percent length change that exceeded those resulting from -200pe were then said to increase the probability of cracking. Linear equations for each mixture group were used in calculating associated percentage length changes for each mixture group. Percent length changes in excess of -0.0342, -0.0478, and -0.0482 were determined to
46


correlate with the cracking of the 100% portland cement mixtures in programs one, two, and three respectively. The mixtures containing SCMs would likely crack if percent length changes occur in excess of -0.0516. Mokarem et al concluded that for 100% portland cement mixtures, 28-day percent length change should be limited to -0.0300 and -0.0400 at 90 days to reduce the risk of cracking due to drying shrinkage. For SCM concrete, percent length change should be limited to -0.0400 at 28-Days and -0.0500 at 90 days.
3.8 Admixtures
Water reducing admixtures are often used in concrete to increase workability while maintaining a low w/cm, resulting in higher strength concrete. A lower w/cm of a concrete mixture will result in reduced drying and plastic shrinkage.
ACI212 Committee Report (ACI 212, 1989) gives detailed information concerning set retarders and set accelerators. Set retarders are sometimes used in bridge deck applications because they offer delayed set times. These retarders allow for continuous placement of bridge decks and make the deck less susceptible to cracking due to deflection of the formwork during placement. The delayed set time is also accompanied by lower temperatures during hydration which help reduce cracking due to thermal stresses (Transportation research circular E-C107, 2006).
47


Xi et al state that there is no definite conclusion on the influence of set
accelerators on bridge deck cracking. The use of retarders increases plastic shrinkage, but decreases the heat of hydration and thermal stresses, resulting in decreased plastic shrinkage cracking (2003).
Shrinkage reducing admixtures (SRAs) are a new product currently undergoing testing and research. They work by reducing the surface tension of the concrete water which reduces internal stresses thus lowering long-term shrinkage. Concrete in the 50% humidity range develop significant capillary stresses which develop into cracks. SRAs reduce these stresses enough to reduce shrinkage cracking and will be incorporated into mixtures used for this research.
48


Chapter 4
Problem Statement
4.1 Statement
Concrete is always going to crack. Even concrete that has been designed accordingly is expected to crack once in service. If concrete cracks during early stages after placement, it immediately begins to degrade the structure. Preventing the early age cracking of concrete is especially important to the CDOT. It is the CDOTs responsibility to maintain a safe network of roads, bridges, and highways throughout the State of Colorado. From public safety to keeping an efficient budget, a durable low cracking potential concrete is very effective in accomplishing both of these objectives. A cracked bridge deck not only diminishes the integrity of the structure but jeopardizes the safety of the travelling public. Substantial damage to the structures integrity begins to occur when cracking in the deck surface allows water to penetrate to the reinforcing steel. The resulting corrosion of steel reinforcement shortens the life span of the bridge and increases maintenance costs while the bridge is in service. These factors are unfavorable, specifically to the department of transportation.
Winter conditions in Colorado create the need for increased de-icing salt on the road surface to ensure the safety of the traveling public. The increased amounts
49


of salt (MgCl2 or magnesium chloride) accelerate the corrosion process when melting snow transports the chlorides through the small cracks to the steel reinforcement.
Research has been underway to investigate several factors contributing to the problems surrounding early age cracking in concrete. The CDOT currently has specifications for low cracking concrete used for bridge decks; Class H and Class HT concrete. Current specifications require fresh and hardened concrete properties of the concrete to fall within a specific range. While the current Class H and HT specifications are an improved approach over previously designed bridge deck concrete, the need for enhancement still exists.
The purpose of this thesis is to design mixtures with material content ranges above and below that of the current specifications. It is believed that the current specifications are creating favorable scenarios for early age cracking. The rate of strength gain, magnitude of ultimate strength, permeability, restrained shrinkage strain, and ffeeze/thaw durability was tested for each of the designed mixtures and their effects on early age cracking examined. Specifically, eleven, low cracking potential, concrete mixtures were designed, batched, and tested for this study.
Fresh and hardened concrete properties were examined and their individual effect on concrete cracking analyzed.
50


The primary benefit gained from this research is that the CDOT will be in a better position to design and construct crack resistant bridge decks and other concrete structures. Results from this study will provide the necessary information to develop more durable concrete bridge decks. This data will allow CDOT to make changes to current specifications for future construction.
Ancillary benefits from this research will include a cost savings to the CDOT. Developing a crack-resistant concrete will benefit the CDOT by providing for longer lasting concrete structures and reducing the annual costs to maintain these pavement structures.
51


Chapter 5
State DOT Survey
5.1 General
A national survey of state Departments of Transportation was conducted with the objective of obtaining additional information that may aid in the improvement of the current CDOT specification for structural bridge deck concrete. A web-based tool called SurveyMonkey.com (http://www.survevmonkey.com/) was used to formulate the questionnaire and analyze the responses. A 38% response rate was obtained for the State DOT survey. Though the response rate was not as high as the study team had hoped, valuable information was gathered from the survey findings. The survey was submitted to state Departments of Transportation (DOT) Materials and bridge engineers. Analysis was performed on the results and aided in the design process of the concrete mixtures created for this research.
5.1.1 Survey Response
Responses were received from 19 of the 50 State DOTS, for a 38% return rate. See Figure 5.1. Multiple states DOTs provided more than one response. Most of the two-respondent states included responses from both the Materials and Bridge
52


Engineer. The survey returned a total of 33 responses; however, only 28 individuals completed the survey.
Figure 5.1 DOT Respondents Map
Multiple responses were obtained from six states: Maryland Transportation Authority, Michigan Department of Transportation, Louisiana Department of Transportation, Tennessee Department of Transportation, Nebraska Department of Roads, and the Arkansas State Highway and Transportation Department.
5.1.2 State DOT Bridge Deck Cracking Problem
A majority of respondents, 95.0%, replied that their state does experience bridge deck cracking. Transverse deck, full width cracking is common and is expected to
53


occur at early ages in many states. In addition, the span type (i.e. continuous spans) with positive and negative moment regions have affected the frequency of cracking.
5.1.3 Potential Causes for Bridge Deck Cracking
The Respondents were asked to choose which of the following choices primarily contributes to bridge deck cracking; placement, curing, rate of strength gain, mixture design, or the use of admixtures. The majority of responses selected curing to be the primary cause of cracking, pertaining to mixture design, placement, rate of strength gain, and use of admixtures ranked from most influential to least influential, respectively. Settlement and early-age thermal cracking are also mentioned as causes for deck cracking.
5.1.4 Rate of Concrete Strength Gain
The Respondents were asked to select at what age their bridge deck concrete typically reaches its ultimate strength; 3, 7, 14, 21, 28, or 56 days. A majority of states, 42.9%, reported achieving ultimate strength at 7 days. Respondents representing 35.7% claim to achieve ultimate strength at 28 days. Of the fourteen responses, no one reported achieving ultimate strength at 3 days of age. The information suggests that it would be beneficial to slow the rate of strength gain for the concrete being designed for this study.
54


5.1.5 AASHTO PP34 Ring Test Usage by State DOTs
A majority of Respondents, 93.8%, replied that their state does not perform AASHTO PP34. Many agree that shrinkage is an important issue contributing to cracking; however do not perform any shrinkage measuring tests. One response reported using the test, but finding little increased strain and zero cracking.
5.1.6 Mixture Design Issues
The respondents had to choose from four choices pertaining to mixture design; water to cementitious material ratio (w/cm), cement content, chemical admixtures, or pozzolans. Half of the respondents report cement content as the major contributor to bridge deck cracking, while 37.5%. report the cause to be the water to cementitious material ratio. Pozzolans were selected only two times and chemical admixtures were not selected by anyone taking the survey.
5.1.7 Mixture Design Modifications Used to Improve Concrete Performance
A common adjustment made by many states is the cement content. Reductions in
j > ->
cement content were mentioned; 6601b/yd to 61 llb/yd and 7091b/yd to 5711b/yd3. An approach taken by the Minnesota DOT is to reduce the permeability of the concrete with lower paste contents and higher percentages of SCM's. Their latest designs involve straight portland cement. The concern of the Minnesota
55


DOT is that the use of supplementary cementitious materials (SCM's) result in lower tensile strengths in the first several days resulting in concrete unable to resist restraint cracking.
5.1.8 Shrinkage-Reducing Admixtures
Only 21.4% of the responses indicated using shrinkage-reducing admixtures in their states bridge deck concrete. The Michigan DOT abandoned a project involving S.R.A.s claiming it repeatedly knocked the air out. An ongoing project currently utilizing S.R.A.s is the Twin Spans Bridge between New Orleans and Slidell. Because the project is ongoing, LADOT has not yet reported whether it was or was not beneficial.
5.1.9 Shrinkage Compensating Cement
Shrinkage compensating cement (Type K, expansive cement) is currently being tested in the United States. Ohio and New York are two of only several states currently utilizing this type of cement. One problem concerning Type K cement is predicting the amount of expansion which will occur. The majority of the respondents, 84.6%, reported having never used shrinkage-compensating cement in their bridge decks.
56


5.1.10 Factors Affecting Cracking (Mixture Design)
Admixtures may contribute to bridge deck cracking. Seven choices were provided for selection as materials commonly found in bridge deck concrete mixtures. The choices were silica fume, Class C fly ash, Class F fly ash, blast-furnace slag, water-reducing admixtures (super-plasticizers), set retarders, or shrinkage-reducing admixtures. Silica fume was chosen by most respondents as the cause of increased cracking. Blast furnace slag and water reducing admixtures were also selected numerous times. Louisiana suspects they are having problems with the compatibility of materials such as cement, admixtures, and fly ash within their mixture. Set retarders and shrinkage-reducing admixtures were chosen least among the provided choices.
5.1.11 Beneficial Factors that Reduce Concrete Cracking
Contrary to the responses presented in 5.1.10, some responses claim that blast furnace slag and water-reducing admixtures proved beneficial in reducing cracking in bridge decks. Some states also claim silica fume to be beneficial against cracking. Iowa DOTs report lower shrinkage when slag and Class C fly ash are used as a ternary blend.
57


5.1.12 Water-to-Cementitious Materials Ratio
Four ranges of w/cm ratio were provided for this question; w/cm < 0.35, 0.35 < w/cm < 0.40, 0.40 < w/cm < 0.45, and w/cm > 0.45. A majority of respondents, 78.6%, selected 0.40 < w/cm < 0.45 as the range for the maximum allowable w/cm for their State DOTs concrete bridge deck mixtures.
5.1.13 Curing Practices
Curing is mentioned by many respondents to contribute significantly to bridge deck cracking. A significant number of respondents, 81.8%, reported changes in their states curing practices of bridge deck concrete. A common response was that an increase in moist-cure (wet cure) times from 7 to 14 days was beneficial. Another is the application of wet burlap within 30 minutes of placement. The Michigan DOT specifies strict fogging, burlap, soaker hose systems for a continuous 7 day wet cure, but reports that enforcement of these specifications is inconsistent. Also, it was noted that monitoring concrete temperature and protection of the concrete during its early plastic state are essential in minimizing concrete cracking.
5.1.14 DOT Survey Conclusion
The results from this survey were utilized when designing concrete mixtures for this study. The information was used by the University of Colorado Denver
58


research team in conjunction with the CDOT. The survey was successful in finding a solid foundation of information from which to begin designing concrete mixtures. In addition, it should be noted that bridge deck cracking is not an isolated phenomenon in Colorado, rather is experienced in most all states.
When designing the survey several options were available for setting up each question. A mistake was made when creating this survey by allowing multiple responses per person per question. The intention was to get a weighted ranking or rating of survey answers. In several questions, a problem occurred when the program attempted to analyze the responses. The answers were lumped into one large category consisting of every response received. As a result, the percentages accompanying any of the analysis reports for these questions are inaccurate.
In summary, several factors such as cement content and concrete curing were noted as being influential factors resulting in concrete cracking of bridge decks for several DOTs. Reduction in the total cementitious content and 14 day cure times are a few adjustments to the mixture design and curing practices made by State DOTs. Further, many DOTs do not perform shrinkage evaluation tests of any kind on their current bridge deck mixtures.
59


Chapter 6
Experimental Design
6.1 Design Plan
6.1.1 Literature Review
A primary objective of this research includes providing the Colorado CDOT with an up-to-date investigation into current research involving the same objectives.
This involved extensive use of the internet to find applicable information about pertinent previous and current research. The review also included close examination of several published theses from various universities, students, and engineers around the world. This information was used in the design process of the eleven concrete design mixtures tested during this research.
6.1.2 Mixture Design Process
Eleven concrete design mixtures were created for research testing. In addition to the DOT survey and literature review, design input was also gathered from meetings with CDOT engineers and other industry professionals interested in the research.
60


6.1.3 Mixture Designs
Ultimately, eleven design mixtures were developed and tested during this research study. See Table 6.1. Four mixtures were designed to reduce the early age accelerated strength gain by limiting the 7-day compressive strength to 3000 psi. This was accomplished by adjusting the w/cm, cementitious content of the mixture, and percent of pozzolan replacement. In addition, the use of coarse-ground cement was incorporated into several mixture designs.
Table 6.1 Mixture Design Matrix
Mix# Mixture ID w/cm Cementitious Content Type of Cement %FA %BFS %SF ADMIX. Air Content (%> Paste Vol.
1 0.38/6.8/FA20/SF5/II 0.38 640 Type II 20 5 6.5 0.28
2 0.42/6.2/FA16/SF3.5/N 0.42 580 Type II 16 3.5 6.5 0.26
3 0.38/6.8/FA20/SF5/G 0.38 640 Class G Oil Well Cement (Coarse Grained Cement) 20 5 6.5 0.28
4 0.42/6.2/FA16/SF3.5/G 0.42 580 Class G Oil Well Cement (Coarse Grained Cement) 16 3.5 6.5 0.26
5 0.44/6.5/FA30/II 0.44 611 Type II 30 6.5 0.29
6 0.44/6.5/FA30/SF5/II 0.44 611 Type II 30 5 6.5 0.29
7 0.44/6.5/BFS50/II 0.44 611 Type II 50 6.5 0.28
8 0.44/6.5/FA30/RET/II 0.44 611 Type II 30 RET. 6.5 0.28
9 0.44/6.5/FA30/SRA/II 0.44 611 Type II 30 SRA. 6.5 0.28
10 0.42/6.0/IKLWA) 0.42 564 Type II 6.5 0.25
11 0.42/6.0/IKNORM.WT. 0.42 564 Type II 6.5 0.25
Key: 0.38/6.8/FA20/SR
w/cm
Cement Content % (sacks) Ash
T ype of Cement
% Silica Fume
Within the eleven concrete mixture designs are two Class H control mixtures, per current CDOT Structural Concrete Specifications. One mixture
61


contains the highest allowable percentage replacement of portland cement with fly ash and silica fume (and lowest allowable w/cm) and the other with the lowest allowable percentage replacement of cement with the same (and highest allowable w/cm). All of the mixtures take into account aggregate content, effective replacement percentages of portland cement with supplementary cementitious materials, chemical admixtures, and varying w/cm. An air-entraining agent (AEA) was used to increase durability of the concrete. Air content within these concrete mixtures was expected to coincide with the required percentages per CDOT structural concrete specifications.
6.1.3.1 Cement Type
Mixtures #1 (0.38-6.8-FA20-SF5-II), and #3 (0.38-6.8-FA20-SF5-G) are CDOT control mixtures and have identical mixture proportions and w/cm equal to 0.38, but Mixture #3 is made using the Type G, oil-well cement which is more coarsely ground than common Type II cement.
Mixture #2 (0.42/6.2/FA16/SF3.5/II) and Mixture #4 (0.42/6.2/FA16/SF3.5/G) are the other CDOT control mixtures but Mixture #4 is again made using the Type G, oil-well cement which is more coarsely ground instead of more common Type II cement.
62


6.1.3.2
Supplementary Cementitious Materials
Mixture #5 (0.44/6.5/FA30/II), Mixture #6 (0.44/6.5/FA30/SF5/II), and Mixture #7 (0.44/6.5/BFS50/II) have the same w/cm (0.44) but each introduces various amounts of cement replacement with supplementary cementitious materials; 30% Class F fly ash alone, 30% Class F fly ash and 5% silica fume, and a mixture containing only 50% blast furnace slag. The 30% replacement of cement with Class F fly ash in Mixture #5 exceeds current allowable CDOT Class H and HT specification replacement percentages of 20%.
6.1.3.3 Chemical Admixtures
Mixture #8 (0.44-6.0-FA30-SRA-II) and Mixture #9 (0.44-6.0-FA30-RET-II) are identical in mixture proportions but each incorporates the use of a chemical admixture. Both mixtures exceed current allowable CDOT Class H and HT specification replacement percentages by having a 30% percent replacement of cement with Class F fly ash. Mixture #8 (0.44-6.0-FA30-SRA-II) utilizes a shrinkage reducing admixture (s.r.a) to help reduce and control the development of shrinkage strain. S.R.A.s are used in the field to help control shrinkage strain development. The s.r.a. was a Master Builders- Tetraguard and the maximum suggested dosage rate of 1.5gal/yd.3 was incorporated. Chemical properties for the shrinkage reducing admixture are provided in Appendix B. Mixture #9 (0.44-6.0-
63


FA30-RET-II) utilizes a set retarder admixture. These admixtures are often used in the field to delay set time when temperatures are high or traffic holds delivery of fresh concrete. The set retarder was a Master Builders- Pozzolith 100XR and an average dosage of 3 ounces per one hundred pounds of cementitious materials in the mixture. Chemical properties for the Pozzolith 100XR can be found in Appendix B.
6.1.3.4 Aggregate Type
Mixture #10 (0.42-6.0-II-L.W.A) is a 100% portland cement mixture made with a substitution of normal weight sand with 2501bs./yd.3 of light weight, fine-aggregate. The aggregate was pre-conditioned (pre-soaked) to a moisture content (m.c.) of approximately 18%. This was an exceptionally high m.c. for aggregate but is done so with the intent of internally curing the concrete. The aggregate releases internal water for use in hydration of cement particles over time. Results are expected to be most significant at 56-days of age. Mixture #11 (0.42-6.0-II-Norm.Wt.) was a control mixture for comparison against the lightweight aggregate concrete mixture. Mixture proportions are identical to Mixture #10 (0.42-6.0-II-L.W.A).
64


6.2 Acquisition of Raw Materials
6.2.1 Cement
Two types of local cement were used in this research study. Holcim Type II Portland cement was supplied by Holcim Inc. and used in the fabrication of several concrete mixtures. In addition, coarse-grained cement supplied by GCC Dacotah Cement from Rapid City, South Dakota, was utilized for two mixtures. This type of cement is a Class G, Oil-well cement. Calcium silicate compounds and other calcium compounds containing iron and aluminum make up the majority of this product. It was expected that concrete mixtures containing this cement develop strength much slower than mixtures containing the Type II cement promoting less shrinkage and more resistance to cracking. The cement reports supplied by the manufacturers for the Holcim Type II and Dacotah Class G Oil-well cement are included in Appendix B. However, the cement compounds, chemical and physical properties and compressive strength properties for the Class G Oilwell cement are shown in Tables 6.2, 6.3, and 6.4 respectively.
Table 6.2 Class G Oilwell Cement Compounds
Dacotah Cement Major Compounds:
3Ca0.Si02 Tricalcium silicate
2CaO.Si02 Dicalcium silicate
3Ca0.Al203 Tricalcium aluminate
4Ca0.Al203.Fe203 Tetracalcium aluminoferrite
CaS04.2H20 Calcium sulfate dehydrate (Gypsum)
65


Table 6.3 Class G Oilwell Cement Chemical and Physical Properties
Chemical Physical
MgO (%) - - 1.2 -
S03 (%) - - 2.2 -
Ignition Loss (%) - 0.8 - -
Equivalent alkalies (%) 0.21 - - -
Insoluble residue (%) - 0.29 - -
C3S - - - 54
C3A - - - 4
Blaine Fineness (m2/kg) 325
Percent Passing No. 325 Mesh, % 84
Free Water, ml 1.4
Table 6.4 Class G Oilwell Cement Compressive Strength Properties
Compressive Strength 8 hours, 100 degree F. at Atm. Press., MPa (psi) N/A
8 hours, 104 degree F. at Atm. Press., MPa (psi) 11.1 (1613)
Pressure Temperature Thickening Time Test Thickening Time, minutes 131
Chemical and physical properties and compressive strength properties for the Holcim Type II cement are shown in Tables 6.5 and 6.6 respectively.
66


Table 6.5 Holcim Type II Cement Chemical and Physical Properties
Chemical Physical
MgO (%) - - 1.2 -
S03 (%) - - 3.2 -
Ignition Loss (%) - 2.4 - -
Equivalent alkalies (%) 0.7 - - -
Insoluble residue (%) - 0.53 - -
C3S - - - 56
C3A - - - 6
Blaine Fineness (m~/kg)
396
Table 6.6 Holcim Type II Cement Compressive Strength Properties
Compressive Strength 3 Day 28.7 (4170)
7 Day 37.0 (5360)
Pressure Temperature Thickening Time Test Thickening Time, minutes 137
6.2.2 Aggregate
Coarse and fine aggregate were obtained from representative sources within Colorado. The UCD Materials Testing Laboratory acquired both the coarse and fine aggregate conforming to the ASTM C33 standard. Bestway Aggregate provided material properties and gradation reports for the aggregate. The aggregate
67


properties and gradation have been checked and verified to meet Class H and HT concrete specifications.
The coarse aggregate meets the ASTM C33 Size Number 57 and 67 gradation requirements. The coarse aggregate was obtained from a source located in Brighton, CO. The fine aggregate meets the ASTM C33 gradation requirement for concrete fine aggregate. Based upon laboratory tests performed by WesTest of Denver, Colorado, this aggregate has a low potential for deleterious alkali-silica behavior. The material properties data for both coarse and fine aggregate are included in Appendix B.
6.2.3 Admixtures
Chemical admixtures used for water-reducing (workability) and air-entrainment, as well as shrinkage reduction and set time were utilized in the design mixtures for this research.
6.2.3.1 High-Range Water Reducing Admixture (H.R.W.R.A.)
A CDOT approved high range water reducing admixture was incorporated into several of the design mixtures. The admixture was manufactured by W.R. Grace-Daracem 19, ASTM C494 Type A and F, and ASTM C1017 Type I. Chemical Properties for the Daracem 19 is provided in Appendix B.
68


6.23.2
Air-Entraining Agent (A.E.A.)
A CDOT approved AEA was utilized for the purposes of air-entraining the concrete design mixtures made for this research. The agent was made by W.R.Grace- Daravair_AT60, ASTM C 260. Chemical properties for the Daravair-AT60 are provided in Appendix B.
6.2.3.3 Shrinkage-Reducing Admixture (S.R.A.)
A CDOT approved shrinkage-reducing admixture was utilized for the purposes of this research. The admixture was supplied by BASF- Master Builders_Tetraguard_AS20. Tetraguard_AS20 product data sheets are included in Appendix B
6.2.3.4 Set Retarder (RET)
A set retarding admixture was utilized for the purposes of this research. The admixture was manufactured by BASF- Master Builders Pozzolith lOOXR. Pozzolith_100XR product data sheets are provided in Appendix B.
6.3 Testing
The mixtures were tested according to ASTM standards for different characteristics occurring from 1 day of age through 56-days of age and beyond. The batching
69


followed ASTM C 192 Standard Practice for Making and Curing Concrete Test Specimens in the Laboratory (AASHTO T 126-97 Making and Curing Concrete Test Specimens in the Laboratory). Both fresh and hardened concrete properties were examined for each mixture batched. The fresh concrete properties that were examined include slump (ASTM C 143, AASHTO T 119), unit weight (ASTM C 138, AASHTO T 121), air content (ASTM C 231, AASHTO T 152), and concrete temperature (ASTM C 1064, AASHTO T 309). Hardened concrete properties that were evaluated in this research included compressive strength (ASTM C 39, AASHTO T 22), unrestrained shrinkage testing (ASTM C 157, AASHTO T 160), restrained ring shrinkage testing (ASTM C 1581, AASHTO PP 34), freeze/thaw durability (ASTM C666, Procedure A, AASHTO 161), and rapid chloride ion penetrability (ASTM C 1202, AASHTO T 227).
In addition to the durability, strength, and permeability testing of the mixtures, the shrinkage strain within the concrete was the primary focus of this research. Throughout the life of the concrete, shrinkage strain results from internal stresses created from the depletion of water. As concrete ages, water is continuously depleted by both the exposed surface evaporation of water and the continuous hydration of the internal cement particles. Restrained ring shrinkage testing allowed for an investigation into the development of allowable strain/stress
70


versus time for each mixture before the concrete cracks. A summary table of test
procedures is shown in Table 6.7.
Table 6.7 Fresh and Hardened Concrete Properties Tests
Fresh Concrete Tests Standard Time of Test
Slump ASTM C 143, AASHTO T 119 At Batching
Unit Weight ASTM C 138, AASHTO T 121 At Batching
Air Content ASTM C 231, AASHTO T 152 At Batching
Temperature ASTM C 1064, AASHTO T 309 At Batching
Hardened Concrete Tests
Compressive Strength ASTM C 39, AASHTO T 22 1,3,7, 28, 56 Days
Rapid Chloride Ion Penetrability ASTM C 1202, AASHTO T 227 28, 56 Days
Durability (F/T Resistance) ASTM C 666, Procedure A AASHTO 161 28 and Subsequent Days
Restrained Shrinkage ASTM C 1581, AASHTO PP34 Until Cracking
6.4 Data Analysis
Resulting test data collected from this research w'as compared and used to provide recommendations for modifications to the current Class H and HT specification, thereby producing a more crack resistant concrete for use as bridge decks by the CDOT.
71


Chapter 7
Experimental Results
7.1 Problems with this Study
7.1.1 Fabrication of Steel Rings
In the summer and fall 2008, the research team experienced difficulty in having the steel rings (per AASHTO PP34) fabricated due to the size of the rings. The research team was able to have four rings fabricated in November 2008. The two additional rings allowed for the testing of two mixtures simultaneously.
7.1.2 Data Acquisition System
A few weeks after batching the second and third scheduled mixtures an issue occurred with the data acquisition system. Immediately after casting each concrete ring (a total of four) they were moved to the humidity controlled curing room and the strain gages connected to the data logger. The original program written to record strain measurements at thirty-minute intervals and used previously would not begin taking measurements. As a result, a program was written immediately which would record strain measurements at two-minute intervals. After a few hours and several attempts, a makeshift program was completed, successfully compiled and downloaded to the data logger. The two minute interval was chosen
72


to accommodate an efficient and timely assurance of a successfully working program. The program completes three cycles (intervals) before it zeros out and begins recording accurate strain measurements. This process took two hours using a program with a specified thirty-minute interval and only eight minutes using a program with a specified two-minute program.
In the hastiness and confusion of writing a new program at this time, a small detail specifying the size of the strain data table was overlooked. The size of the table recording the data was left at a default record size of 11,060, instead of being changed to an unlimited range. As a result of the program recording strain measurements at two minute intervals the table filled up before the test was completed; stopping recording data before the rings cracked. As a result the second and third scheduled mixtures (0.38-6.8-FA20-SF5-II and 0.42-6.2-FA16-SF3.5-II, respectively) were re-batched.
7.1.3 Re-Batch of Mixture #1 (0.38-6.8-FA20-SF5-II) and Mixture #2 (0.42-6.2-FA 16-S F3.5-II)
Upon weighing out the aggregate for the second batch of mixtures one and two (0.38-6.8-FA20-SF5-II and 0.42-6.2-FA16-SF3.5-II) the restock of aggregate used for the CDOT research was on order. There was enough aggregate to satisfy the batch weights, however, the remaining portion of coarse-aggregate contained
73


noticeably more fines in its make-up than previously observed. The remaining portion of the coarse-aggregate supply was churned to ensure consistency and uniformity before removal from the aggregate bins. In the interest of time and progression of research, the mixtures were batched using the near-end of the University of Colorado at Denvers Materials Testing Laboratory coarse-aggregate supply. The size of these re-batches were reduced to a size which would produce only enough concrete to make two restrained shrinkage rings (per mixture) and enough 4 x 8 cylinders to test compressive strength. The mixture proportions for batch two of each mixture were identical to those used in the first batch. The compressive strength cylinders from each of the re-batched mixtures are only being used to compare the mixtures consistency between the first and second batch. The comparison between the mixtures (first and second batches) is being provided to verify mixture consistency since all other specimens fabricated during the first batching of these mixtures (e.g. unrestrained shrinkage beams and freeze/thaw beams) were continued and used to collect research data for these mixtures. The AASHTO PP34 test ran for a few weeks before the error occurred and the rings needed to be re-fabricated. All other test specimens fabricated during the first batch (e.g. unrestrained shrinkage and freeze/thaw) of mixtures one and two (0.38-6.8-FA20-SF5-II and 0.42-6.2-FA16-SF3.5-II, respectively) did not experience any problems and were used for research data for their respective mixtures. The
74


restrained shrinkage specimens fabricated during the second batching of the mixtures (0.38-6.8-FA20-SF5-II and 0.42-6.2-FA16-SF3.5-II, respectively) were used to collect shrinkage strain through test end; cracking.
7.2 Fresh Concrete Properties
Fresh concrete tests included temperature, air content, unit weight, and slump. Fresh concrete properties for the eleven design mixtures are listed in Table 7.1. Table 7.1 Fresh concrete properties
Mixture Identification Slump Air Content Unit Weight Ambient Temp. Concrete Temp.
(in.) (%) (lbs./ft.3) (F) (F)
0.38/6.8/FA20/SF5/II 3.0 5.5 142.4 59 58
0.42/6.2/FA16/SF3.5/II 4.5 8.0 134.2 56 58
0.38/6.8/FA20/SF5/G 3.5 3.4 147.8 59 62
0.42/6.2/FA16/SF3.5/G 5.0 9.5 137.2 62 60
0.44/6.5/FA30/I1 8.0 4.5 143.8 62 59
0.44/6.5/FA30/SF5/II 6.5 9.0 135.8 72 69
0.44/6.5/BFS50/II 3.5 3.5 146.4 72 68
0.44-6.0-FA30-SRA-II 3.0 2.8 147.4 74 71
0.44-6.0-FA30-RET-II 3.0 7.5 141.4 72 71
0.42-6.0-11 (L.W.A) 2.5 7.5 138.6 72 72
0.42-6.0-II (Normal Wt.) 2.0 7.5 143.0 66 69
7.2.1 Slump
Current Class H and HT specifications do not specify a slump value. For adequate workability the desired slump was 3.5 inches (8.89cm). Although some values fall below the target, all eleven design mixtures achieved sufficient
75


workability to form test specimens. The use of a High Range Water Reducing Admixture (HRWRA) and Air Entraining Admixture (AEA) was required to obtain the needed workability and durability sought for this research.
7.2.1.1 Cement Type
Mixture #1 (0.38-6.8-FA20-SF5-II) vs. Mixture #3 (0.38/6.8/FA20/SF5/G), and Mixture #2 (0.42-6.2-FA16-SF3.5-II) vs. Mixture #4 (0.42-6.2-FA16-SF3.5-G) are CDOT Class H control mixtures which examined the effect of coarse-ground cement versus the specified Type II cement. When comparing the slump values between the mixtures made using Type G, coarse-ground cement and Type II cement, the coarse ground cement concrete mixtures achieved an increased slump average of 0.5 inch (1,27cm) over the Type II cement concrete mixtures.
Mixtures #2 (0.42-6.2-FA16-SF3.5-II) and #4 (0.42-6.2-FA16-SF3.5-G) have a w/cm equal to 0.42 and required less HRWRA than Mixtures #1 (0.38-6.8-FA20-SF5-II) and #3 (0.38/6.8/FA20/SF5/G), which both had w/cm equal to 0.38. Mixture #4 (0.42-6.2-FA 16-SF3.5-G) resulted in a slightly higher slump value than those mixtures with a w/cm equal to 0.38.
76


7.2.1.2 Supplementary Cementitious Materials
Fly ash is known to increase workability. Figure 7.1 shows Mixture #5 (0.44/6.5/FA30/II) with an increased w/cm of 0.44 and a 30% replacement percentage of cement with fly ash had significantly increased workability. In fact, Mixture #5 achieved the largest slump (8.0 in., 20.32 cm). This slump is higher than what is usually desirable in the field. Mixture #6 (0.44/6.5/FA30/SF5/II) is the same mixture but with a 5% replacement of cement with silica fume. Silica fume was expected to decrease workability and did so by 1.5 inches (3.81 cm). The 50% blast furnace slag mixture decreased workability significantly from the comparison mixtures #5 and #6 (5in. and 3.5in. respectively.
7.2.1.3 Chemical Admixtures
Mixtures #5 (0.44/6.5/FA30/II), Mixture #6 (0.44/6.5/FA30/SF5/II), Mixture #7 (0.44/6.5/BFS50/II), Mixture #8 (0.44-6.0-FA30-SRA-II), and Mixture #9 (0.44-6.0-FA30-RET-II) have a w/cm equal to 0.44 and did not require any H.R.W.R.A. for workability. The chemical admixtures used in Mixtures # 8 and #9 did not result in increased workability except due to the added moisture in the concrete.
77


7.2.1.4 Aggregate Type
Mixture #10 (0.42-6.0-II-L.W.A) and Mixture #11 (0.42-6.0-II-Norm.Wt.) with a w/cm equal to 0.42 and required very little H.R.W.R.A... An advantage Mixture #10 (0.42-6.0-II-L.W.A) has over the other mixtures is the use of pre-soaked light weight aggregate (L.W.A.). The additional water in the presoaked aggregate helped to increase slump (0.5in., 1.27 cm).
Each of the eleven mixtures attained adequate workability to mold all necessary test samples. Slump test results are shown in Figure 7.1.

,*£y

<0
Q>




,%v .vSb- r 1> T>
£
&
/ /

^ *

? <§? # s,.
4>' $r ^ < f

Jo'

Mixture Identification

Figure 7.1 Slump Test Results (ASTM C 143, AASHTO T 119)
78


7.2.2 Air Content
The use of an AEA was incorporated for all eleven mixtures. Current Class H and HT specifications require air content between 5% 8%. The air content of the research mixtures varied throughout the research. The W.R. Grace air-entraining agent specifies a dosage of 1 fluid ounce per 100 pounds of cementitious materials. This dosage was measured correctly but resulted in random air contents. Previous research using the same dosage rate of AEA has repeatedly proven accurate air content results. The research team believes the error in air content to be caused by excessive cement replacement percentages with cementitious materials (Fly Ash) which caused unforeseen resulting air contents. Although trial batches were made to test the interaction between the various admixtures, the research team believes the interaction between chemical admixtures and high cementitious replacement percentages caused the design mixtures to have variable air contents.
Mixture #3 (0.38-6.8-FA20-SF5-G) was batched first and the exact dosage was used for air content designed to be 6.5%. Mixture #3 (3.4%) is lower than the design of 6.5% by a margin of error equal to 48%. As a result, AEA dosages were re-evaluated for more accuracy. Mixture #1 and #2 were batched next. The AEA dosage was adjusted before batching Mixtures #1 and #2.
All of the mixtures using H.R.W.R.A. required an amount different from the design to achieve adequate workability. The two mixtures having lower w/cm
79


equal to 0.38 both required more than the design amount of HRWRA. As a result, the extended mixing time sometimes required to incorporate the H.R.W.R.A. uniformly into the mixture essentially deflated the concrete, releasing the entrained air. This is typically the case with the mixtures having lower air contents than 6.5%.
Air contents also varied due to experimental replacement percentages of cement with supplementary cementitious materials and the use of chemical admixtures. These experimental mixtures sometimes had unexpected admixture interactions which were not anticipated during design. Air content values are provided in Figure 7.2.
80


Air Content (%)


O
fP

&







&


jF
JP


ro'
k£
^ /' r\/N

*
**



<5- 0
Mixture Identification






ty
Figure 7.2 Air Content (ASTM C 231, AASHTO T 152)
7.2.3 Unit Weight
The unit weight of each mixture was determined at batching per ASTM C 138. The unit weight is the weight of a unit volume of concrete (Equation 2).
Urn,Weigh! = Mightof
Concrete Volume(ft. )
Equation 2
81


Full Text

PAGE 1

EVALUA nON OF CRACK RESISTANT CONCRETE FOR COLORADO BRIDGE DECKS by Robert Wayne Cavaliero B.S. Civil Engineering University of Colorado at Denver 2007 A thesis submitted to the University of Colorado at Denver/Health Sciences Center in partial fulfillment of the requirements for the degree of Master of Science Structural Engineering Civil Engineering

PAGE 2

2009 This thesis for the Master of Science degree by Robert W. Cavaliero has been approved by Stephan A. Durham ChengYu Li Date

PAGE 3

Cavaliero, Robert Wayne (MS, Structural, Civil Engineering Department) Evaluation of Crack Resistant Concrete for Colorado Bridge Decks Thesis directed by Dr. Stephan A Durham ABSTRACT Cracking of reinforced concrete is a problem that has baffled maintenance and bridge engineers since it was first developed. Cracking allows water and contaminates to enter the structure and corrode the reinforcing steel. A concrete structure that is constructed with a concrete that is less susceptible to cracking will last longer and be more maintenance free than one that is susceptible to cracking. The Colorado Department of Transportation (CDOT) has developed a Class Hand HT concrete that is meant to fill the requirement of a crack resistant concrete. Though the current specification is an improvement over the Class D concrete that is commonly used there are still some problems with cracking. The basis for this research includes the design of over ten concrete mixtures in an effort to create a more crack resistant concrete than the current COOT Class Hand HT concrete specification. Cracking is known to be the result of many factors including shrinkage which contributes in large amounts. The concrete mixtures designed for this research were designed with water to cementitious material's amounts and cement replacement percentages both above and below the current

PAGE 4

specifications The design approach was intended to investigate the effect of individual and multiple cementitious materials replacement on several tests of fresh and hardened concrete; restrained and unrestrained shrinkage strain compressive strength rate of strength gain, freeze / thaw durability permeability. A national survey was conducted and offered to all fifty state DOT bridge and/or materials engineers. They were asked about current and past research involving crack resistant concrete as well as comments regarding their state DOT specifications currently used for bridge decks. The results of this survey are used to provide valuable feedback to the Research Team for use in improving the current COOT Class Hand HT specifications. The information was also taken into consideration during the design process of the concrete mixtures for this research A more crack resistant concrete mixture was developed through this study which was designed for better performance in Colorado's bridge decks including less maintenance and improved durability over the life of the structure. This abstract accurately represents the content of the candidate's recommend its publication. Signed

PAGE 5

DEDICA nON PAGE I dedicate this thesis to my friends and family for their patience and understanding throughout this process You can never understand how much I appreciate your understanding for the distance created in our relationships and your patience in waiting for me to complete such a project. To my parents for their never-ending support and inspiration to become what I never thought was possible.

PAGE 6

ACKNOWLEDGMENT I sincerely thank my academic advisor, Dr. Stephan Durham, who secured this research for the benefit of both the University of Colorado at Denver and the Colorado Department of Transportation. His continued advisement has taught me and so many others. Also, I would like to thank Dr. Kevin Rens and Dr. Li for participating on my thesis committee. Thank you to the Colorado Department of Transportation for a great experience. My sincere Thanks to Rui Liu, Tom Thuis, Randy Ray, Driss Majdoub, Adam Kardos, Logan Young and the many others who contributed to this work. I would like to thank Mr. Ken Stevens with Campbell Scientific Laboratories for his many hours of help fine-tuning the data logger program Thank you to all those who donated time and material's to complete this research; Dan Bentz and Jesse (BestWay Aggregate), Matt Finger and Kevin Kane (Holcim Cement), Brandon Cooke (BASF), Tom Green and Bill Hart (W.R.Grace), George Grygar and Miles Dee (txi), David Darwin and JoAnne Browning (University of Kansas), and Jason Weiss (Purdue University). Additionally, I would like to thank the faculty and staff of the University of Colorado at Denver, Civil Engineering Department for their support and guidance throughout my educational career at UCD.

PAGE 7

TABLE OF CONTENTS Figures ............... ............................ ................................. .................................... xii Tables ................................................................................................................. xviii Chapter 2. 2.1 2.1.1 2.1.2 2.1.2.1 2.1.2.2 2.2 2.2.1 2.3 2.3.1 2.3.2 2.3.3 2.3.4 2.4 2.4.1 3. 3.1 3.2 3.3 3.3.1 3.3.2 3.3.2.1 3.4 3.4.1 3.4.2 3.4.3 3.4.4 3.4 5 Introduction ......................................................................................... 1 Concrete ... .. .................................................................................... ...... 1 Problematic Cracking in Concrete Bridges .............................................. 1 Background ...... ...................................................................................... 4 Colorado Department of Transportation .................................................. 4 Research Interest ....................................... ..... ..... ....... ...... ..... ... ............... 4 Current Specifications .............................................................................. 6 Class H Specifications .............................................................................. 6 Class HT Specifications ......................... ........................... ........ .............. 7 Cracking in Concrete ..................................................... ............ ..... ......... 8 Importance of Cracking in Concrete .................. .... ........ ............ .............. 8 Causes of Cracking in Concrete .... ...... ........ ............................................. 9 Internal Stresses ... .............................. ................................................. ... 9 External Stresses and Normal Use Degradation .................................... 10 Restraint ................................................................................................. 10 Shrinkage Strain ... ......... ...... ....... ........ ......... ....... ....... .... ..................... 10 Research Objectives ........ ...................................................................... 12 Objectives of Investigation .................................................................... 12 Literature Review ........................................................... ... .......... .......... 13 Preface ................................................................................................. 13 Curing ............ ....... .... ................................... ...... ........ .... .................... 13 Concrete Shrinkage ................................................................................ 14 Effect of Restraint on Shrinkage ............................................................ 14 Effect of Curing on Shrinkage .............................................................. 15 Internal Curing Using Light Weight Aggregate ..................................... 20 Design Mixture Factors Affecting Cracking in Concrete ...................... 23 Silica Fume ............................................................................................. 23 Fly Ash ................................................................................................. 25 Water to Cementitious Materials Ratio 28 Cement Content ............ ......................................................................... 29 Cement Type ............................. ....... ........ ................. ..... .... .... ... ......... 30

PAGE 8

3.4.5.1 3.4.5.2 3.4.5.3 3.4.6 3.4.7 3.4.7.1 3.4.7.2 3.5 3.6 3.6.1 3.7 3.8 4. 4.1 5. 5.1 5.1.1 5.1.2 5.1.3 5.1.4 5.1.5 5.1.6 5.1.7 5.1.8 5 1.9 5.1.10 5.1.11 5.1.12 5.1.13 5.1.14 6. 6.1 6.1.1 6.1.2 6.1.3 6.1.3.1 General Effects of Cement Fineness ...................................................... 30 Coarse-Ground Cement .......................................................................... 32 Shrinkage Compensating Cements ........................................................ 33 Aggregate Content ................................................................................. 35 Aggregate Composition ......................................................................... 37 General ................................................................................................. 37 Aggregate Composition and Water to Cementitious Materials Content ............................. .... .................................................. 37 Unrestrained Shrinkage Test .................................................................. 39 AASHTO PP34 / ASTM C 1581 ........................................................... 42 Restrained Ring Shrinkage Test.. ............................................ ... ............ 44 Length Change .............. ........ ......... ... .... .......... ..... ..... .......................... 46 Admixtures ............................................................................................. 47 Problem Statement ................................................................................. 49 Statement ................................................................... ................. ...... ..... 49 State DOT Survey .................................................................................. 52 General ............................................... ......... .... .................................... 52 Survey Response .................................................................................... 52 State DOT Bridge Deck Cracking Problem ........................................... 53 Potential Causes for Bridge Deck Cracking ........................................... 54 Rate of Concrete Strength Gain ............................................................. 54 AASHTO PP34 Ring Test Usage by State DOTs .................................. 55 Mixture Design Issues ............................................................................ 55 Mixture Design Modifications Used to Improve Concrete Performance ............... ............ ................ ........................ .................... 55 Shrinkage-Reducing Admixtures ........................................................... 56 Shrinkage Compensating Cement .......................................................... 56 Factors Affecting Cracking (Mixture Design) ....................................... 57 Beneficial Factors that Reduce Concrete Cracking ............................ ... 57 Water-to-Cementitious Materials Ratio ................................................. 58 Curing Practices ............ ............ ....................... .......... ............ ...... ... 58 DOT Survey Conclusion .... .................................................................... 58 Experimental Design ................................................... .......... ............... 60 Design Plan ............................................................................................ 60 Literature Review ................................................................................... 60 Mixture Design Process .............................................. .......................... 60 Mixture Designs ..................................................................................... 61 Cement Type .......................................................................................... 62

PAGE 9

6.1.3.2 6.1.3.3 6.1.3.4 6.2 6.2.1 6.2.2 6.2 3 6.2.3.1 6.2.3.2 6.2.3.3 6.2 3.4 6.3 6.4 7. 7.1 7.1.1 7 1.2 7.1.3 7.2 7 .2.1 7.2.1.1 7 2.1.2 7.2.1.3 7 2.1.4 7.2.2 7.2.3 7.2.4 7.4 7.4.1 7.4.1.1 7.4.1.2 7.4.1.3 Supplementary Cementitious Materials .... .. .... ............ ... ........... ........... 63 Chemical Admixtures ............... .......................... ........................... ... .... 63 Aggregate Type ............................. ... ........ .... ............ ..... ... ...... ...... .. ....... 64 Acquisition of Raw Materials ................................................................ 65 Cement ....................................... ........... ................. ................. ... ....... 65 Aggregate .... ... ...... ... ...................................................... ........................ 67 Admixtures ..................... .... ..... .............................................................. 68 High-Range Water Reducing Admixture (H.R.W.R.A.) ....................... 68 Air-Entraining Agent (A.E.A.) ................... .. .... ...... ... .............. ............ ... 69 Shrinkage-Reducing Admixture (S.R.A.) .............................................. 69 Set Retarder (RET) ........ ........ .................. ........ .. .................... ........... ...... 69 Testing ............................................. ................................................... 69 Data Analysis ......... ... .................... ...... ........ ... ................. ............. ... .... .. 71 Experimental Results ..................... ..... ....... .... .. ............... .... ................... 72 Problems with this Study ................................................................ ....... 72 Fabrication of Steel Rings .......................... ........... ............... ...... ......... 72 Data Acquisition System .......................................................... ............. 72 Re-Batch of Mixture # 1 (0.38-6.8-F A20-SF5and Mixture #2 (0.42-6.2-FAI6-SF3.5-II) ...... .......... ........................ ........ 73 Fresh Concrete Properties .... .... ... ........................ .. ....... .................. ..... 75 Slump ................................................................................................. 75 Cement Type ...... ............. ................. .. ..... ....... ... ................ ... ...... ..... ..... 76 Supplementary Cementitious Materials ...... .... ... ................................... 77 Chemical Admixtures ... ................... ..... ..... ......... ............... ............. .... ... 77 Aggregate Type ................................................... .. ................. .. ............. 78 Air Content ... .. .................. ........................................................... ... ...... .. 79 Unit Weight ............. ............................................................... ...... .. ..... 81 Concrete Temperature ...... ................... .............. .... .... ..... .... ........ ... ..... 83 Hardened Concrete Tests ...................................................................... 85 Compressive Strength ................................................ .. ......... ................ 85 Mixtures Having Inadequate 56-Day Strength ......... ... .......... ... ............. 87 Normalization of Compressive Strength ............ ................................... 89 Comparison of Mixture #1 (0.38-6.8-FA20-SF5-II) and Mixture #2 (0.42-6.2-FAI6-SF3.5-II), Batch One and Two .................. 92 7.4.1.4 Early-Age Compressive Strength ................ ..... .. .................................... 96 7.4.1.4.1 Cement Type ........ .................................... .... ......................................... 98 7.4.1.4.2 Supplementary Cementitious Materials .... .. ............... ........................ 102 7.4.1.4 3 Chemical Admixtures .............................. ........................... .... ....... ... ... 106 7.4.1.4.4 Aggregate Type ....... ...... ........................... .... ..... ......... ......................... 107 7.4.1.5 Ultimate Strength (28-day and 56-Day) .... .... ...... .... ............... ............. 109

PAGE 10

7.4.1.5.1 Cement Type ........................................................................................ 109 7.4.1.5.2 Supplementary Cementitious Materials ............................................... 112 7.4.1.5.3 Chemical Admixtures ........................................................................... 114 7.4.1.5.4 Aggregate Type .................................................................................... 115 7.4.2 Permeability ......................................................................................... 117 7.4.2.1 General ............................................................................................... 117 7.4.2.2 Rapid Chloride Ion Penetrability Test.. ................................................ 119 7.4.2.2.1 Cement Type ........................................................................................ 122 7.4.2.2.2 Supplementary Cementitious Materials ............................................... 127 7.4.2.2.3 Chemical Admixtures ................... ....................................................... 130 7.4.2.2.4 Aggregate Type .................................................................................... 132 7.4.3 Durability ............................................................................................. 134 7.4.3.1 General ............................................................................................... 134 7.4.3.2 Durability Analysis .............................................................................. 155 7.4.3.2.1 Cement Type ........................................................................................ 155 7.4.3.2.2 Supplementary Cementitious Materials ............................................... 159 7.4.3.2.3 Chemical Admixtures ........................................................................... 160 7.4.3.2.4 Aggregate Type ................................. ................................................. 16 1 7.4.4 Restrained Shrinkage Strain ................................................................. 162 7.4.4.1 General ............................................................................................... 162 7.4.4.2 Strain Analysis ..................................................................................... 166 7.4.4.2.1 Cement Type ........................................................................................ 166 7.4.4.2.2 Supplementary Cementitious Materials ............................................... 175 7.4.4.2.3 Chemical Admixtures ........................................................................... 178 7.4.4 2.4 Aggregate Type .................................................................................... 183 7.4.4.3 Paste Content (Volume) ....................................................................... 189 8. Conclusions and Recommendations .................................................... 196 8 1 Fresh Concrete Properties .................................................................... 196 8.1.1 Slump ............................................................................................... 196 8.1.2 Air Content ................. ................................. ..... .......................... ......... 197 8.1.3 Unit Weight .......................................................................................... 197 8.1.4 Temperature ......................................................................................... 198 8.2 Mixture Design Properties ................................................................... 198 8.2.1 General ............................................................................................... 198 8.3 Recommendations ................................................................................ 202

PAGE 11

Appendix A. Concrete Design Mixtures .............................. ...... ..... ... ... ........ ............... 206 Materials Product Data ............................................................................... 217 C. DOT Survey ....................................... .................. ....................... ............ 234 D. Photographs of Cracked Restrained Ring Shrinkage Test Specimens ....... 239 Bibliography ............... .... .... ....... . ...... ............... ........... ....... ... ... ............ 203

PAGE 12

FIGURES Figure 5.1 DOT Respondents Map ................................................................................ 53 7.1 Slump Test Results, (ASTM C 143, AASHTO T 119) ................................ 78 7 2 Air Content, (ASTM C 231, AASHTO T 152) ............................................ 81 7.3 Unit Weight (ASTM C 138, AASHTO T 121) vs. Air Content (ASTM C 231, AASHTO T 152 .................................................... 83 7.4 Concrete Temperature, (ASTM C 1064, AASHTO T 309) ......................... 84 7.5 Photograph of Compressive Strength Failure (ASTM C 39, AASHTO T 22) ..................................................................... 86 7.6 56-Day Compressive Strength (ASTM C 39, AASHTO T 22) ................... 89 7.7 56-Day Compressive Strength vs. 56-Day Compressive Strength (Normalized for Air Content), (ASTM C 39, AASHTO T 22) ...................................................................................... 91 7.8 28-Day Compressive Strength, CDOT Control Mixture #1 (0.38-6.8-FA20-SF5-11), Batch One vs. Batch Two, (ASTM C 39, AASHTO T 22) ..................................................................... 93 7.9 28-Day Compressive Strength, CDOT Control Mixture #2 (0.42-6.2-FA16-SF3.5-II), Batch One vs. Batch Two, (ASTM C 39, AASHTO T 22) .................................................................................. 94 7.10 Early-Age Compressive Strength, (ASTM C 39, AASHTO T 22) .............. 97 7.11 Early-Age Compressive Strength, CDOT Control Mixture #1 (O.38-6.8-FA20-SF5-II) (Type II Cement) and Mixture #3 (0.38-6.8-F A20-SF5-G) (Type G, Coarse-Ground Cement), (ASTM C 39, AASHTO T 22) ........................... 99 7.12 Early-Age Compressive Strength, CDOT Control Mixture #2 (0.42-6.2-FA16-SF3.5-1I) (Type II Cement) and Mixture #4 (0.42-6.2-FA 16-SF3.5-G) (Type G, Coarse-Ground Cement), (ASTM C 39, AASHTO T 22) ................................................................... 100 7.13 Early-Age Compressive Strength, Mixture #5 (0.44-6.5F A30-II), Mixture #6 (0.44-6.5-F A30-SF5-II), and Mixture #7 (0.44-6.5-BFS50-II),(ASTM C 39, AASHTO T 22) .............. 104 7.14 Early-Age Compressive Strength, Mixture #8 (0.44-6.0F A30-SRA-IJ) (Shrinkage Reducing Admixture) and Mixture #9 (0.44-6 0-FA30-RET-II) (Set Retarding Admixture) (ASTM C 39, AASHTO T 22) ...................... 106 7.15 Early-Age Compressive Strength Mixture #10 (0.42-6.0-Il-Light Weight Aggregate) and Mixture #11 (0.42-6.0-II-Normal Weight Aggregate), (ASTM C 39 AASHTO T 22) .................................. 108

PAGE 13

7.16 Compressive Strength, CDOT Control Mixture # 1 (0.38-6.8-F A20-SF5-II) (Type II Cement) and Mixture #3 (0.38-6 8-F A20-SF5-G) (Type G Coarse-Ground Cement) (ASTM C 39, AASHTO T 22) ............... .............. .......... ......................... 110 7 .17 Compressive Strength CDOT Control Mixture #2 (Type II Cement) and Mixture #4 (0.42-6.2-FAI6-SF3.5-G) (Type G, Coarse-Ground Cement) (ASTM C 39 AASHTO T 22) .......... 111 7.18 Compressive Strength, Mixture #5 (0.44-6.5-FA30-II) Mixture #6 (0.44-6.5-F A30-SF5-II) and Mixture #7 (0.44-6.5-BFS50-Il), (ASTM C 39 AASHTO T 22) ........ ..... .... .... ...... ....... .. ........ ...... .. .......... .... 113 7.19 Compressive Strength, Mixture #8 (0.44-6.0-F A30-SRA-II) (Shrinkage Reducing Admixture) and Mixture #9 (0.44-6.0-FA30-RET-II ) (Set Retarding Admixture), (ASTM C 39 AASHTO T 22) .... ....... .... ....... ..................... .... .... .... .... ...... .... .. .... 114 7.20 Compressive Strength Mixture #10 (0.42-6.0-II-Light Weight Aggregate) and Mixture #11 (0.42-6.0-II-Normal Weight Aggregate), (ASTM C 39 AASHTO T 22) .......... ..... ........ ......... ..... ........ 116 7.21 Photograph of R.C.I.P. Test Setup ............... ......................... ................... 118 7.22 Rapid Chloride Ion Penetrability Test Results (ASTM C 1202, AASHTO T 227) ... ...... ........ ......... ..... ........ ................ ...... ......... ............ 121 7.23 56-Day Rapid Chloride Ion Penetrability Test Results (ASTM C 1202 AASHTO T 227) .......... ........... ..... ........... .. ...... ..... .... .... .... ...... .... 122 7.24 28-Day and 56-Day Rapid Chloride-Ion Penetrability Test Results, CDOT Control Mixture #1 (0. 38-6.8-FA20-SF5-II) (Type II Cement) and Mixture #3 (0.38 -6.8-F A20-SF5 G) (T y pe G Coarse-Ground Cement) (Permeability ASTM C 1202, AASHTO T 227) ................... ......................................... ........ .... 124 7.25 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results CDOT Control Mixture #2 (0.42-6 2-FA16-SF3.5-II) (Type II Cem e nt) and Mixture #4 (0.42-6.2F A16-SF3 5-G) (Type G Coarse-Ground Cement) (ASTM C 1202 AASHTO T 227) ... 126 7.26 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results Mixture #5 (0.44-6.5-FA30-II) Mixture #6 (0.44-6.5-F A30-SF5-II) and Mixture #7 (0.44-6.5-BFS50-II) (Permeability, ASTM C 1202 AASHTO T 227) ...... .... .............. .... ..... .... 129 7.27 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results Mixture #5 (0.44-6.5-FA30-II), Mixture #6 (0.44-6.5-F A30-SF5-II) and Mixture #7 (0.44-6.S-BFS50-II), (Permeability ASTM C 1202 AASHTO T 227) ............................ ............... ....... ........ ............ 130

PAGE 14

7.28 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results, Mixture #8 (0.44-6.0-FA30-SRA-II) (Shrinkage Reducing Admixture) and Mixture #9 (0.44-6.0-FA30-RET-II) (Set Retarding Admixture), (Permeability, ASTM C 1202, AASHTO T 227) ..................................... 131 7.29 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results Mixture #8 (0.44-6.0-FA30-SRA-II) (Shrinkage Reducing Admixture) and Mixture #9 (0.44-6.0-F A30-RET II) (Set Retarding Admixture), (Permeability ASTM C 1202 AASHTO T 227 ..................................... 132 7.30 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results, Mixture #10 (0.42-6.0-II-Light Weight Aggregate) and Mixture #11 (0.42-6 0-II-Normal Weight Aggregate) (Permeability, ASTM C 1202, AASHTO T 227) ........................................................................................ 134 7.31 Photograph of Freeze/Thaw Chamber (ASTM C 666, Procedure A) ........ 136 7.32 Photograph of Durability Testing Apparatus (ASTM C 666, Procedure A) ...................................................................... ............... 137 7.33 Photograph of Durability Testing Apparatus (ASTM C 666, Procedure A) .............................................................................................. 138 7.34 Durability Factor and Air Content CDOT Control Mixture #1 (0.38-6 8-FA20-SF5-II) and COOT Control Mixture #2 (0.42-6.2-F A 16-SF3.5-II) ........................................................................... 156 7.35 Durability Factor and Air Content, CDOT Control Mixture #1 (0.38-6.8-FA20-SF5-II) and Mixture #3 (0.38-6.8-FA20-SF5-G) ............. 157 7.36 Durability Factor and Air Content, COOT Control Mixture #2 (0.42/6.2/F A 16/SF3.51II) and Mixture #4 (0.42/6.2/F AI6/SF3.5/G) ........ 158 7.37 Durability Factor and Air Content Mixture #5 (0.44/6.5/F A301II), Mixture #6 (0.44/6.5/F A30/SF51II), and Mixture #7 (0.44/6.5/BFS50/U) ................................... ..... .............................. ..... ..... 160 7.38 Durability Factor and Air Content Mixture #8 (0.44-6.0-FA30-SRA-II) ............................................................................ 162 7.39 Photograph of Restrained Ring Shrinkage Specimen (ASTM C 1581, AASHTO PP34) ........................................................................... 163 7.40 Photograph of Restrained Ring Shrinkage Specimen, (ASTM C 1581, AASHTO PP34) ........................................................................... 165 7.41 Restrained Shrinkage Strain, CDOT Control Mixture # 1 (0.38-6.8-F A20-SF5-II) and Mixture #3 (0.38-6.8-F A20-SF5G), (ASTM C 1581, AASHTO PP34) .............................................................. 167 7.42 of 56-Day Strength Achieved at Respective Age, Mixture #1 (0.38-6.8-FA20-SF5-II) and Mixture #3 (0.38-6.8-FA20SF5-G), (ASTM C 39, AASHTO T 22) ..................................................... 169

PAGE 15

7.43 of Ultimate Strain Achieved, CDOT Control Mixture #1 (0.38-6.8-F A20-SF5-II) and Mixture #3 (0 38-6.8-FA20SF5-G), (ASTM C 1581, AASHTO PP34) ........................... .................... 170 7.44 of 56-Day Strength Achieved at Respective Age, CDOT Control Mixture #2 (0.42/6 2/FAI6/SF3.5/11) and Mixture #4 (0.42/6.2/FAI6/SF3.5/G), (ASTM C 39, AASHTO T 22) ............... ..... .............................. ..... ............... ..... ........ 171 7.45 of Ultimate Strain Achieved at Respective Age, CDOT Control Mixture #2 (0.42/6.2/FAI6/SF3.5/I1) and Mixture #4 (0.42/6.2/FAI6 / SF3.5/G), (ASTM C 1581, AASHTO PP34) ............ ...... 173 7.46 Restrained Shrinkage Strain, CDOT Control Mixture #2 (0.42 / 6.2/FAI6/SF3.5/I1) and Mixture #4 (0.42/6 2/FAI6/SF3.5 / G), (ASTM C 1581, AASHTO PP34) ........................................ ................... 174 7.47 Restrained Shrinkage Strain Mixture #5 (0.44/6.5/FA301II), Mixture #6 (0.44/6.5/F A30/SF51II) and Mixture #7 (0.44/6.5/BFS501II), (ASTM C 1581, AASHTO PP34) ............... ............ 176 7.48 of Ultimate Strain Achieved at Respective Age, Mixture #5 (0.44 / 6.5 / F A301II), Mixture #6 (0.44/6.5/F A30/SF5/I1), and Mixture #7 (ASTM C 1581, AASHTO PP34) ............ ..... .... ..... ..... ...... ............................... .... ........... ... 178 7.49 Restrained Shrinkage Strain, Mixture #8 (0.44-6.0-FA30-SRA-II) and Mixture #9 (0.44-6.0-F A30-RET -II), (ASTM C 1581, AASHTO PP34) .... ................. ..... .... ... ........... .... ..... .......... .... .......... ....... 180 7.50 of 56-Day Strength Achieved, Mixture #8 (0.44-6.0-FA30-SRA-II) and Mixture #9 (0.44-6.0-FA30-RET-II) (ASTM C 39, AASHTO T 22) ............................. ..................................... 181 7.51 of Ultimate Strain Achieved, Mixture #8 (0.44-6.0-FA30-SRA-II) and Mixture #9 (0.44-6.0-FA30-RET-II), (ASTM C 1581, AASHTO PP34) .... ................ ........................................................ ........ 183 7 52 of 28-Day Strength Achieved, Mixture #10 (0.42-6.0-1I-L.W.A) and Mixture #11 (0.42-6.0-II-Norm Wt.) (ASTM C 39, AASHTO T 22) .......................................................................................... 185 7.53 % of Ultimate Strain Achieved, Mixture #10 (0.42-6.0-II-L.W.A) and Mixture #11 (0.42-6.0-Il-Norm.Wt.), (ASTM C 1581, AASHTO PP34) ................... ..................................................................... 187 7.54 Restrained Shrinkage Strain, Mixture #10 (0.42-6.0-II-L.W A) and Mixture #11 (0.42-6.0-II-Norm.Wt.), (ASTM C 1581, AASHTO PP34) ................................................... . ............ .......... 188

PAGE 16

7.55 % of Ultimate Strain Achieved vs. Paste Content (29 vs. 25%), Mixture 5 (0.44/6.5/F A301II) and Mixture #6 (0.44/6.5/F A30/SF5/II) vs. Mixture #8 (0.44/6.0/F A30/SRAIII) and Mixture #9 (0.44/6.0 / FA30/RETIII) (ASTM C 1581, AASHTO PP34) .. ....................................................................................... 193 7.56 % Ultimate Strain Achieved vs. Paste Content (28 vs. 26%) Mixture # I (0.38/6.8 /F A20/SF51II) and Mixture #2 (0.42/6.2/FAI6/SF3.5/IJ) respectively (ASTM C 1581, AASHTO PP34) ............. ........ ............. .... ................................................... 195 Appendix B Figure B.l Fine Aggregate Gradation (ASTM C 33) ...... .................. ........................... 217 B.2 Coarse Aggregate Gradation (ASTM C 33) ...... ............. .......................... 218 B.3 WesTest Aggregate Test Results ... ..... .......... .. ..... .... ............ .. ................... 219 B.4 Holcim Type II Cement Properties ............................................................ 222 B.5 Boral Material Technologies, Class F Fly Ash (ASTM C 618 T Report) ......................................................................................... 223 B.6 W.R Grace, Daracem 19, High Range Water Reducing Admixture,Product Data ............................ ............................................... 226 B.7 W.R.Grace, Daravair AT60, Air-Entraining Admixture (ASTM C 260), Product Data ..... ................. ... ........... ........... ...... ............ 228 B.8 BASF Tetraguard AS20, Shrinkage-Reducing Admixture, Product Data ................................................................................ .... .. ... ..... 230 B.9 BASF Pozzolith 100XR Set-Retarder Admixture, Product Data ..................................... ........ ... ....................... ...................... 232 Appendix C Figure DOT Survey, Questions #1 and #2 ............. ..... .... ................ ..................... 234 C.2 DOT Survey Questions #3, #4, #5, and #6 ..... ......................................... 235 C.3 DOT Survey, Questions #7, #8, #9, and #10 ....... ........... ..... ...... ............... 236 C.4 DOT Survey Questions # # 12, and # 13 ........................ .................. .... 237 C.5 DOT Survey Respondent Contact Information ......................................... 238

PAGE 17

Appendix D Figure 0.1 Photograph of Mixture #3 (0.38-6.8-F A20-SFS-G), Ring 1 Restrained Ring Shrinkage Test Specimen ........... ... .... .... ....................... 239 0.2 Photograph of Mixture #2 (0.42-6.2-F A 16-SF3.S-II) Ring2 Restrained Ring Shrinkage Test Specimen ............. ............................... 240 D 3 Photograph of Mixture #4 (0.42-6.2-FA 16-SF3.S-G), Ring2 Restrained Ring Shrinkage Test Specimen ... ..... .................................... 241 0.4 Photograph of Mixture #S (0.44-6.S-FA30-II) Ringl Restrained Ring Shrinkage Test Specimen ........ ... ............... ... . ........ 242 D.S Photograph of Mixture #6 (0.44-6.S-F A30-SFS-II) Ring2 Restrained Ring Shrinkage Test Specimen .......... ... ..... ...................... 243 0.6 Photograph of Mixture #9 (0.44-6.0-FA30-RET-II) Ringl Restrained Ring Shrinkage Test Specimen ............... ............ ........... 244

PAGE 18

TABLES Table 2.1 Class H and Class HT mixture specifications ................................................ 4 6.1 Mixture Design Matrix ....... ................... .................... ..... ............................ 61 6.2 Class G Oilwell Cement Compounds ......... ..... ............................. ..... .......... 65 6.3 Class G Oilwell Cement Chemical and Physical Properties ........................ 66 6.4 Class G Oilwell Cement Compressive Strength Properties ...... ............. .... 66 6.5 Holcim Type II Cement Chemical and Physical Properties ......................... 67 6.6 Holcim Type II Cement Compressive Strength Properties ...... .................. 67 6.7 Fresh and Hardened Concrete Properties Tests ........................................... 71 7.1 Fresh Concrete Properties ........ ................................................................... 75 7.2 Compressive Strength (ASTM C 39, AASHTO T 22) ........... ....... ....... .... ... 87 7.3 Normalized Compressive Strength ............ ..... ..... ....... ................... ... ........... 90 7.4 Permeability Rating per Coulombs Passed ..... .......................... ...... ......... 119 7.5 Rapid Chloride Ion Penetrability Results (ASTM C 1202, AASHTO T 227) ........................................................................................ 120 7 6 Mixture #1 (0.38-6.8-FA20-SF5-II) freeze/thaw results ........................... 139 7.7 Mixture #2 (0.42-6.2-F A 16-SF3 .5-11), freeze/thaw results ........................ 140 7 8 Mixture #3 (0.38-6.8-FA20-SF5-G), freeze/thaw results ............... ... ..... ... 141 7.9 Mixture #4 (0.42-6.2-F A 16-SF3.5-G) freeze/thaw results ............. ......... 141 7.10 Mixture #5 (0.44-6.5-FA30-II) freeze/thaw results .................................. 142 7.11 Mixture #6 (0.44-6.5-FA30-SF5-II) freeze/thaw results ........................... 142 7.12 Mixture #7 (0.44-6.5-BFS50-II), freeze/thaw results ................................ 143 7.13 Mixture #8 (0.44-6.0-FA30-SRA-II) freeze/thaw results ......................... 143 7.14 Mixture #9 (0.44-6.0-FA30-RET-II), freezelthaw results ......................... 144 7.15 Mixture #10 (0.42-6.0-II-light weight aggregate), freeze/thaw result ....... 144 7.16 Mixture #11 (0.42-6.0-II-normal weight aggregate), freeze/thaw results .. 144 7.17 Mixture #1 (0.38-6.8-FA20-SF5-II), relative modulus of elasticity .......... 147 7.18 Mixture #2 (0.42-6.2-F A 16-SF3.5-II), relative modulus of elasticity ....... 148 7.19 Mixture #3 (0.38-6.8-FA20-SF5-G), relative modulus of elasticity ....... ... 149 7.20 Mixture #4 (0.42-6.2-FA 16-SF3.5-G), relative modulus of elasticity ....... 150 7.21 Mixture #5 (0.44-6.5-FA30-II), relative modulus of elasticity .................. 150 7.22 Mixture #6 (0.44-6.5-F A30-SF5-II), relative modulus of elasticity .......... 151 7.23 Mixture #7 (0.44-6.5-BFS50-II), relative modulus of elasticity ................ 151 7.24 Mixture #8 (0.44-6.0-F A30-SRA-II) relative modulus of elasticity ......... 152 7.25 Mixture #9 (0.44-6.0-FA30-RET-II), relative modulus of elasticity ..... .... 152 7.26 Mixture #10 (0.42-6.0-II-light weight aggregate), relative modulus of elasticity ..... .... ............ ............................ ............................. 153

PAGE 19

7.27 Mixture #11 (0.42-6.0-II-normal weight aggregate), relative modulus of elasticity .................................................................................. lS3 7.28 Durability Factors ....................................................................................... lS4 7.29 Design Mixture Properties ......................................................................... 190 8.1 Compressive Strength, Permeability, and Restrained Shrinkage Test Results ............................................................................... 200 8.2 Comparison Between Study Mixtures and Class Hand HT Specification Requirements ........................................................................ 201 Appendix A Figure A.l Concrete Design Spreadsheet, Mixture # 1 (0.38-6.8-FA20-SFS-II) .......... 206 A.2 Concrete Design Spreadsheet, Mixture #2 (0.42/6.2/FA16/SF3.S/Il) ...... 207 A.3 Concrete Design Spreadsheet, Mixture #3 (0.38-6.8-FA20-SFS-G) ......... 208 A.4 Concrete Design Spreadsheet, Mixture #4 (0.42/6.2/F A 16/SF3.S/G) ....... 209 A.S Concrete Design Spreadsheet, Mixture #S (0.44/6.S/F A30/Il) .................. 210 A.6 Concrete Design Spreadsheet, Mixture #6 (0.44/6.S/FA30/SFS/II) .......... 211 A.7 Concrete Design Spreadsheet, Mixture #7 (0.44/6.S/BFSSO/Il) ................ 212 A.8 Concrete Design Spreadsheet, Mixture #8 (0.44-6.0-FA30-SRA-II) ........ 213 A.9 Concrete Design Spreadsheet, Mixture #9 (0.44-6.0-F A30-RET -II) ........ 214 A.10 Concrete Design Spreadsheet, Mixture #10 (0.42-6.0-II-L.W.A ............ 215 A.11 Concrete Design Spreadsheet, Mixture #11 (0.42-6.0-II-Norm.Wt.) ........ 216

PAGE 20

Within the past five years, the Colorado Department of Transportation (CDOT) has experienced a continued problem with cracking of bridge decks. In 2003, CDOT implemented concrete mixture designs Class H and Class HT into the CDOT Class Hand HT were developed to provide crack resistant concrete structures and were intended to be used in the construction of bridges and other concrete structures (CDOT, 2005). Recently, the CDOT has noticed cracking in several bridge decks using these concrete specifications. Cracking in residential or city roadways will allow for delayed repair. Traffic can swerve to avoid cracks or potholes for months if they get large enough. However, bridge decks are suspended above rivers, mountain valleys, deep crevasses, and sometimes, other roadways carrying the traveling pUblic. If a bridge deck develops sufficient cracking as to damage its structural integrity a failure could result and be disastrous. The result of such large structural failures is of immense possibilities, most of which result in human injury or death. In addition

PAGE 21

to safety CDOT and other state DOT's are interested in low-cracking potential concrete in an effort to reduce maintenance costs and delays to the motoring public. Ultimately the primary objective is to improve the performance of concrete bridge decks in Colorado by minimizing cracking potential of the concrete mixtures used in them. Cracking in reinforced concrete structures allows water and contaminants to migrate inside the structure where it can cause deterioration of the reinforcing steel as well as the surrounding concrete matrix. Water that is able to penetrate through the bridge superstructure can also cause damage to the substructure and affect bridge aesthetics The de-icing chemicals used during inclement weather to provide safe driving conditions in combination with air and water accelerates the corrosion of reinforcing steel (rust or oxidation). The existing bond between the concrete and the steel diminishes as the corrosion process progresses, jeopardizing the integrity of the structure. When a bridge is in service and experiences cracking, naturally the cracks grow with time. This allows for more water and de-icing chemicals to enter the deck and degrade the reinforcing steel creating the need for replacement or repair earlier than normal. This perpetuation of bridge deterioration requires costly and labor-intensive repair. To minimize the amount of cracking and reduce maintenance costs, Class H and HT concrete mixtures were analyzed in this Thesis to ensure the concrete meets the 2

PAGE 22

expectations of the CDOT. Further, additional mixtures were evaluated for their effectiveness in reducing cracking in concrete structures. To accomplish this, eleven concrete mixtures were designed with low-cracking potential as the primary objective. The results of this study and recommendations to the CDOT are included in this thesis. 3

PAGE 23

In 2003, the Colorado Department of Transportation (CDOT) revised their to include two new classes of structural concrete. Class H and Class HT concrete were included into the standard specifications as a crack resistant concrete. These concretes are currently used in the construction of bridges and other concrete structures. Class H concrete is used for concrete bridge decks without a topping slab and waterproofing membrane [Xi et. aI, 2003]. Class HT concrete is used as a top layer for exposed concrete bridge decks. The design criterion for each of these concrete classes is shown below. Class H and Class HT mixture specifications 3 3 3 M M 4

PAGE 24

A study on Colorado bridge decks was published in March 2003 [Xi et aI, 2003]. The objectives of this study were twofold. First, the extent and causes for bridge deck cracking was investigated. Secondly, concrete material properties, construction practices, and design specifications where examined as to possible causes for bridge deck cracking. A literature review within this study concluded that cracking in early age bridge decks is a result of material, design, construction, and environment. High early age shrinkage was found to be a major cause for this cracking problem. In addition, the structural design had a direct role in cracking as well. Cracks were typically noticed above girders and piers. Placement and curing can have a significant role in cracking, primarily plastic shrinkage cracking. Recommendations regarding materials, design factors, and construction practices were included in the Final Report. Cement and silica fume content, water/cement ratio, and the rate of strength gain were key recommendations regarding materials included in the report. Recently, the CDOT has discovered a number of bridge decks throughout the state constructed with Class H and Class HT concrete that exhibit cracking. is suspected that the rate of strength gain for these concrete mixtures may in part be a contributing factor to this cracking. Several bridge decks have obtained the 28-day compressive strength within three days. Other factors that influence cracking

PAGE 25

include: types and amount of aggregate, cement content and type, water / cement ratio, and air content. These are discussed in more detail in Chapter 3 of this thesis. Colorado's harsh weather conditions make it essential for the states bridge decks to have strict performance and mixture specifications. Early age cracking of bridge decks can decrease the life of the structure and increase maintenance costs immensely. Traffic safety and an efficient use of materials and labor are of great interest to the CDOT. For this reason, the CDOT has requested an investigation into developing a new specification for concrete with low cracking potential. Class H concrete is used for bare concrete bridge decks with no waterproofing membrane. Below is a summary of current CDOT Class Hand HT specifications. 56-day compressive strength of 4500 Ibs.lin.2; Required air content of 5% 8% are required; Water-to-Cementitious Ratio (w/cm) ranging from 0.38 0.42; An approved water reducing admixture; A minimum of 55 percent AASHTO M 43 size No. 67 coarse aggregate by weight of total aggregate; 6

PAGE 26

Laboratory trial mixture must not exceed permeability of 2000 coulombs at 56-days of age (ASTM C 1202) and must not exhibit a crack at or before 14 days in the cracking tendency test (AASHTO PP 34). 2.1.2.2 Class Specifications The CDOT Class Hand HT concrete have identical specifications and are used for bare concrete bridge decks that will not receive a waterproofing membrane. The difference between the Class Hand HT lies in that Class HT concrete is used as the top layer of the bare bridge deck. The specifications for the CDOT Class HT concrete are summarized below: 56-day compressive strength of 4500 lbs'/in?; Air content of 5% 8% are required; W / cm ranging from 0.38 0.42; An approved water reducing admixture; Must have a minimum of 50 percent AASHTO M 43 size No 7 or No.8 coarse aggregate by weight of total aggregate Laboratory trial mixture must not exceed permeability of2000 coulombs at 56-days (ASTM C 1202) and must not exhibit a crack at or before 14 days in the cracking tendency test (AASHTO PP 34). 7

PAGE 27

Concrete is known to be weak in tension. In design concrete beams are assumed to have zero tensile strength These tensile stresses are fairly low when compared to those experienced by reinforced bridge decks or beams, which spans are restrained between two or more supporting structures. Individual lanes of bridge decks are sometimes placed while others on the bridge remain open for service. For many reasons, bridge decks experience movement (deflection) during daily traffic and thermal expansion which can contribute to the concrete cracking. The earlier the concrete deck cracks the faster the rate of deterioration and need for repair. As a result, the concrete must be more durable and designed to have characteristics that will be advantageous during early ages and in this environment. A decrease in early age cracking will delay the development of corrosion on the reinforcing steel, decreasing its permeability and increasing the structures durability. Cracking in reinforced concrete structures allows water and contaminants to migrate inside the structure where it can cause deterioration of the reinforcing steel as well as the surrounding concrete matrix. Water that is able to penetrate through the bridge superstructure can also cause damage to the substructure and affect bridge aesthetics. 8

PAGE 28

To minimize the amount of cracking and reduce maintenance costs, Class H and HT concrete mixtures was analyzed in this study to ensure the concrete meets the expectations of the CDOT. Further additional mixtures were evaluated for their effectiveness to eliminate or at least reduce cracking in concrete structures. Concrete cracks as the result of any number of factors. Internal stresses within the concrete are the primary cause of early-age cracking. Internal stresses develop depending upon the heat of hydration, the rate of strength gain, ultimate 28-day and 56-day compressive strength, cement content, percent replacement of cement with supplementary cementitious materials (SCMs) and (Equation 1); w = Equation 1 Additionally, the use of chemical admixtures is necessary to create various desirable characteristics of the mixture. These characteristics include reducing shrinkage, delayed set time and air content. All of which can impact the magnitude and rate of development of internal stresses and cause cracking. 9

PAGE 29

Daily, cyclical service loading is a major cause of cracking in concrete bridge decks. These stresses are unavoidable as the Colorado weather, temperature fluctuation, and traveling vehicles gradually degrade the roadways and deck surfaces. Restraint has long been an issue regarding bridge deck cracking. Deck slabs are restrained against movement at joints and internally around steel reinforcement. As concrete expands thermally or shrinkage occurs, the restraint against movement will result in cracking. Expansion joints in bridges help to alleviate cracking due to these stresses. Shrinkage strain is a major cause of early age cracking in concrete and the primary focus for this research. Multiple types of shrinkage exist and are all detrimental to the life of the concrete. As water leaves the cement paste matrix, the volume of cement paste begins to shrink and is termed 'shrinkage.'

PAGE 30

Drying shrinkage represents the strain caused by the loss of water from concrete. This type of shrinkage results in surface cracking (map-cracking) and causes the surface of the bridge deck to deteriorate at a much faster rate. A type of drying shrinkage is termed autogenous shrinkage, which occurs as the internal water is gradually depleted during the continued hydration of cement particles over the life of the concrete. Regardless of the type of shrinkage, the volume of the cement paste has a tendency to shrink as the water dissipates. Shrinkage begins to occur immediately after the concrete sets, as surface water begins to evaporate and with the continued hydration of cement particles. The voids in the concrete once occupied by water are then left empty. The volume shrinkage that attempts to occur within the rigid cement paste matrix creates internal stresses within the concrete. These stresses induce a strain on the concrete that results in early age cracking. This research utilizes the AASHTO P34 Restrained Ring Shrinkage Test to measure these shrinkage strains versus time. The primary objective of this research is to design, batch, and test a minimum of ten concrete mixtures to examine various aspects of concrete mixtures and their influence on cracking. Specifically, this research aims to develop a concrete mixture that is more resistant to cracking than the current Class Hand HT specification. A more detailed explanation and understanding of all the tests

PAGE 31

perfonned for this research is included in the literature review in Chapter 3 of this thesis. 2.4 Research Objectives 2.4.1 Objectives of Investigation The primary objectives of this study are to design a more crack resistant concrete for use in Colorado's bridge decks. Upon completion, recommendations will be provided the Colorado Department of Transportation for a new low cracking concrete specification or to augment current Class Hand HT concrete specifications. 12

PAGE 32

3.1 Preface Chapter 3 Literature Review This literature review does not examine the effects of super-structure design on concrete bridge deck cracking. Construction practices such as curing, finishing, time of placement (ambient temperature), and consolidation playa major role in bridge deck cracking. This thesis investigates the effect of mixture design factors which influence bridge deck cracking. Curing practices are discussed herein only to emphasize its importance in the practice of placing concrete. 3.2 Curing Curing is not the focus of this thesis; however, curing is essential to producing quality concrete. Curing is the method used to reduce the evaporation of water immediately after placement and is required to promote continued hydration of the cement, thereby increasing the concrete's compressive strength and overall durability. The effect of curing cannot be neglected in practice. Furthermore, the effect of curing on compressive strength and shrinkage cannot be disregarded. All of the research examined for this literature review discusses the importance of 13

PAGE 33

adequate curing. Internal curing is the only method of curing pertaining to the scope of this research and is discussed in further detail in Section 3.4 7.3.3 3.3 Shrinkage is a major cause of cracking in concrete bridge decks. When cement is hydrated and water evaporates, internal stresses develop and volume shrinkage of the concrete occurs, autogenous shrinkage and drying shrinkage, respectively. The hardened concrete attempts to resist these stresses and cracks as a result. A concrete mixture design may combat shrinkage by adjusting the quantity of anyone or multiple materials used in making concrete. A literature review was conducted on several available studies involving cracking in concrete bridge decks. The research information reviewed was built upon in an effort to efficiently provide the CDOT with revised and more durable bridge deck mixture designs. 3.3.1 Restraint has long been known to cause bridge deck cracking. As a concrete bridge deck dries and moisture evaporates, it experiences a volume decrease termed shrinkage. According to Krauss and Rogalla (1996), the amount of shrinkage depends primarily on the paste content and water content. Reinforcement and the bridge superstructure components such as girders provide restraint against 14

PAGE 34

shrinkage, resulting in tensile stresses that cause the concrete to crack (1996). Restrained ring shrinkage tests (AASHTO PP34, ASTM C 1581) allow researchers to conduct a relative comparison of the micro strain associated with different mixture materials at the point of cracking due to restraint in a controlled environment. Cracking is indicated as the point when the strain in the steel ring suddenly decreases. The exposed surface of the concrete ring makes inspection for cracks simple although several mixtures did not exhibit visible surface cracking after the drop in micro strain occurred. The standard specifically states this test is not accurately applicable to field practice or exposed structures. The restrained ring test is not applicable to expansive cements or concrete having a nominal maximum aggregate size (NMAS) greater than 13 mm (0.50 in.). If any of the concrete rings do not crack during the test period, the rate of tensile strength stress development at the time the test is terminated provides a basis for comparison of the materials (ASTM C 1581). Although multiple methods of curing are not included in the scope of this research, the method used to cure concrete is essential to its characteristics such as durability, rate of strength gain, ultimate strength, freeze/thaw resistance, and appearance. Cement paste will never completely hydrate when the w/c ratio is below 0.42. A 15

PAGE 35

layer of C-S-H builds up on the largest grains of cement and hinders the hydration process. Curing helps ensure as much hydration as possible occurs and at a reasonable cost (Mindess, Young and Darwin, 2003) After meeting with the CDOT it was discovered that training on the importance of curing techniques was non-existent leaving a huge opportunity for project error. A survey of other state DOT's further strengthened the widespread belief suggesting curing practices are a major cause, perhaps the primary cause, of transverse deck cracking. Krauss and Rogalla performed their own survey of existing DOT's fifteen years ago (1993) and received many of the same responses concerning curing. They discovered many curing practices were being used in different states depending upon the job but that no standard curing practice existed for bridge decks. Practices ranged from allowing only membrane or curing compounds to requiring long-term wet curing using curing compounds and in many cases the contractor was given the liberty to choose the method. Krauss and Rogalla suggest the latter practice will most likely result in problems with the concrete. Typically, the contractor would choose the cheapest method to save money but the cheapest method is not typically the most effective one for the job. Babaei and Hawkins (1987) point out that fogging or evaporation retarding films substantially reduce early plastic deck cracking if applied immediately after strike-off of the concrete. In addition Babaei and Hawkins suggest applying wet burlap as soon as possible. This method results in 16

PAGE 36

fewer smaller cracks than curing compounds; delayed water curing even increases cracking. Krauss and Rogalla (1993) reported high cement content concrete to be most affected by curing. Concrete with a ratio equal to O.SO and cement content of278 kglm3 (470 Ib/yd3 ) that was wet cured for 60 days experienced little change in time to first cracking of the ring in the restrained ring shrinkage test. When the ratio was lowered to 0.3S, cement content increased to SOl kglm3 (846 Ib/yd\ and curing remained the same, time to first cracking of the ring increased from 11.7 to 21.0 days. Mindess, Young, and Darwin (2003) suggest the duration of and the maximum temperature reached by the cement paste plays a major role in cracking. They report pastes which achieve elevated temperatures during curing experience reduced irreversible shrinkage with no effect on reversible shrinkage. A paste exposed to 6SoC (lS0F) reduces irreversible shrinkage by 66.67% and total shrinkage by 33.33%. This reduction is attributed to the large proportion of the capillary porosity having formed as macro pores, resulting in a reduced micro porosity of C-S-H. The effective reduction in shrinkage is a function of the duration of exposure time to higher temperatures. According to Mindess, Young, and Darwin the exposure time necessary to reduce shrinkage can be relatively short and is often less than the total curing time (2003). 17

PAGE 37

Wet curing techniques such as quickly applying wet burlap, water ponding, or continuous water misting are all beneficial curing methods that reduce cracking by reducing the evaporation rate of water in concrete. High performance and high cement content concrete only have a small amount of mixture water to evaporate. Wet curing not only slows down the rate of water evaporation but cools the concrete simultaneously. This results in lower thermal stresses that develop due to the heat of hydration (Krauss and Rogalla, 1993). Mixed opinions exist as to what is the ideal curing method. Krauss and Rogalla suggest the immediate use of windbreaks and wet curing the concrete. Curing should consist of misting, curing compound, and wet burlap. The minimum curing period is 7 days, ideally 14 days, when the evaporation rate exceeds 1 kg/m2/hr (0.2 Ib/ft2/hr) for normal concrete and 0.5 kg/m2/hr (0.1 Ib/ft2/hr) for concrete susceptible to early-age cracking due to low ratios. They report that exposure to high temperatures after the curing period is complete can also help to reduce irreversible shrinkage. Most researchers agree that a standardized method of curing is needed and should be instated by AASHTO. Oeshpande et al (2007) examined the effect of the curing length on air entrained concrete made with both Type IIII and Type II coarse ground cement. Concrete made with Type IIII cement exhibited significantly increased shrinkage when comparing curing durations at different periods of time beyond initial drying.

PAGE 38

At 30 days beyond initial drying the shrinkage of concrete cured for 3, 7, 14 and 28-days were (micro strain), 375, 340 and respectively. As the curing period increased, the free shrinkage decreased. This trend continued through measurements taken up to 365 days past initial drying. At 365 days past drying the largest difference in shrinkage strain occurred between concrete cured for 3 and 7 days 690 and respectively. Differences in strain were small between concrete cured for 7 and 14 days at 525 and respectively. Air-entrained concrete made with Type II coarse ground cement exhibited a similar trend; shrinkage decreased with increased curing periods. At 30 days past drying, concrete cured for 3, 7, 14, and 28-days experienced free shrinkage micro strains of 250 205, 110 and respectively Concrete cured for 3 days experienced slightly more shrinkage than concrete cured for 7 days until approximately 75 days past drying. After that point the difference in free shrinkage results were relatively small. At 180 days past drying a difference of approximately existed between the concrete cured for 3 to 7 days and those cured for 14 to 28-days. is apparent from the results that an extended curing period creates a more durable concrete for both Type VII and Type II coarse ground cement concrete. It is clear that the ultimate shrinkage of concrete made with Type 1111 cement is significantly higher than Type II coarse ground cement concrete at all ages. Free shrinkage measurements were taken at intervals of 30 180, and 365 days past drying on 19

PAGE 39

concrete cured for 3 days and a difference of225, 240, and 30011, respectively, existed between the Type VII and Type II coarse ground cement. The research performed by Deshpande et el (2007) clearly shows the advantage of using Type II coarse ground cement over a Type lIII cement when the effect of curing periods on shrinkage are being considered. The use of new presoaked lightweight aggregate (L W A) in high performance concrete (HCC) is becoming more common. The aggregate is said to internally cure as a result of it being soaked before batching and contributes to the hydration process instead of absorbing water from the concrete mixture. This approach uses aggregate made of porous expanded shale, sufficient to provide effective internal curing in order to reduce self-desiccation and autogenous shrinkage cracking. Cusson and Hoogeveen conducted research (2006) at the Canadian Institute for Research and Construction examining high performance concrete made with Type I portland cement and replacements of sand by the light weight aggregate. A control mixture was designed with a cement-sand-stone ratio of 1 :2:2 and equal to 0.34. It is noted that the water used to pre-soak the L W A was accounted for in the calculation of the and remained constant for all of the concretes examined. This requirement was said to have made the evaluation of the internal curing

PAGE 40

effectiveness more severe than if additional water had been used to soak the aggregate. The three batches substituted nonnal weight sand with 6, 12, and 20% pre-soaked L W A and a fourth control mixture substituting 0% L W A. One large concrete prism 200 x 200 x 1000 mm (8 x 8 x 40 in.) was cast for each mixture with reinforcement and used a setup attaching strain gauges to the steel in order to detennine the restrained shrinkage. A second concrete prism of the same size was cast from each mixture without reinforcement and used for unrestrained shrinkage testing. This prism was cast with thennal couples and relative humidity (RH) sensors (measuring self-desiccation) implanted within the fresh concrete. Compressive strength and splitting tensile strength tests were also perfonned on 100 x 200 mm (4 x 8 in.) cylinders. The 20% L W A concrete experienced reduced drying due to the internal curing. The RH of the control specimen reduced from 100% at set time to 98% after 2 days and 96% after 7 days. The RH of the 20% L W A concrete reduced to 98% after 2 days and 94% at 7 days. The control test specimen had a 7 day compressive strength of 50MPa (7252 Ibs/in2) versus the 20% L W A concrete of 57MPa (8267 Ibs/in 2). Cusson and Hoogeveen attribute this to the improved hydration of the pre-soaked L W A. Free shrinkage test results prove that as the L W A content increased in the concrete mixtures the autogenous shrinkage decreased. The 0, 6, 12, and 20% L W A concretes experienced strains of 2521-l (micro strain), 210, 112, and 461-l respectively at 2 days of age. After

PAGE 41

restrained shrinkage tests were performed the stress/strength curves were normalized. This was done so the comparison could be made between the various curves corresponding to different concretes, which require different degrees of restraint during testing. Restraints varied from a low 0.9% for the 0% and 6% L W A concrete in order to avoid failure, to a high restraint of 1.1 for the 20% L W A concrete, having the loading system slightly pulling on the prism. The replacement of sand with L W A increased the modulus of elasticity (MOE) considerably. At 3 to 4 days of age, the MOE was several thousand MPa higher for the 20% LWA concrete than the control (0% LWA) concrete. The 7 day splitting tensile strengths were measured to be 4.lMPa (595 Ibslin2), 4.8MPa (696Ibs/in2), 4.5MPa (653 Ibslin\ and 4.2MPa (609 Ibslin2) for the 0%, 6%, 12% and 20% L W A concretes respectively. The maximum stress/strength ratio achieved by the 20% L W A concrete was 50% after nearly 3 days. These results illustrate the L W A to be extremely beneficial in reducing cracking. Cusson and Hoogeveen's research shows how effective internal curing is against shrinkage and tensile stress in concrete, especially high performance concrete. Their results prove the effect of L W A sand replacement on strain and stress reductions. Their data indicates that a 25% L W A concrete could possibly eliminate autogenous shrinkage and tensile stress. Significant swelling did occur in the 20% L W A concrete. As a result, it is 22

PAGE 42

not recommended to use more than a 25% L W A concrete because of the possibility of excess swelling (Cusson and Hoogeveen 2006). Substitution of cement with silica fume produces a denser concrete matrix. results in a more rapid rate of hydration which is accompanied by a higher heat of hydration and increased early strength development (Transportation Research Circular E-C 1 07 2006) A higher heat of hydration results in higher thermal stresses and reduced bleeding, making concrete more prone to plastic shrinkage (Xi et aI, 2003) Another study by Bissonnette, Pierre, and Pigeon (1999) also claims silica fume is not beneficial in concrete for reducing cracking One of their research programs compared two concrete mixtures with w / cm equal to 0.33. One of the mixtures contained 15% silica fume substitution for portland cement. Restrained ring shrinkage tests were performed and the silica fume concrete produced an additional 300 micro strains at 4 days over the 100% portland cement concrete. Bissonnette et al concluded that the presence of silica fume in concrete results in an increase in long term shrinkage. However the resulting early age increase in shrinkage leads to significant cracking because the tensile strength is so low at early ages (1999). 23

PAGE 43

Whiting, Detweiler, and Lagergren (2000) also researched the effect of silica fume on concrete shrinkage in full depth decks and concrete overlays. Full depth mixtures used lower cementitious material contents and air contents with higher than the overlay design mixtures. Silica fume substitution ranged from 0 to 12 percent of the total cementitious material weight and for overlays ranged from 0.30 to 0.35; full-depth decks ranged from 0.35 to 0.45. Unrestrained drying shrinkage tests AASHTO T 160 (ASTM 157) were performed on three 75 x 75 x 285 mm (3 x 3 x 11.25 in.) prisms molded for each mixture. The unrestrained test specimens were cured in lime saturated water; full-depth mixtures were cured for 7 days and overlay mixtures only 3 days. They were then moved to a controlled relative humidity of 50% and a temperature of23 C (73 F). Restrained shrinkage tests were performed per ASTM C 1581 (AASHTO PP34) on a 75 mm (3 in ) thick, 150 mm (6 in.) high concrete ring around the outside of a 19 mm (0.75 in.) thick steel cylinder having an outside diameter of300 mm (11.75 in.). The restrained ring specimens were cured for periods of 1 and 7 days, intending to represent both the worst and best field curing practices. A data acquisition system wired to four strain gauges that were attached (90 offset) around the inside of the steel ring measured the strains at thirty minute time increments. Their results show the presence of silica fume to have little effect on long term shrinkage (450 days). Early age shrinkage (4 days) was higher for 24

PAGE 44

concrete mixtures with silica fume, versus the control mixtures made without. At this age, results consistently show an increase in shrinkage with increased silica fume content. Lower concrete mixtures (0.36) demonstrated less shrinkage when made with a constant replacement of silica fume (1.8%) than concrete with a higher (0.43). The lower the the smaller the cement content relative to the mixture. The smaller the amount of cement in the mixture means the smaller the amount of cement paste formed that is available to shrink. Whiting et al point out that the two mixtures having of 0.36 and 0.43 had paste volumes of25.2 and 27.5% respectively. is also noted that small variations in may greatly influence shrinkage in concrete. The silica fume specimens cured for one day cracked sooner than control specimens. For specimens cured 7 days, the silica fume in the concrete significantly reduced time to first crack. Whiting et al suggest not exceeding 6% silica fume replacement of portland cement because it begins to have an adverse negative effect on shrinkage and cracking. Research concerning the replacement of portland cement with fly ash in a concrete mixture has returned contradicting results. Class F and class C fly ash replacement is a very effective method of slowing the rate of C-S-H growth. reduces early age strength gain and early concrete temperatures and still achieves the same 25

PAGE 45

ultimate strength (Xi et aI, 2003). High volumes of fly ash substitution for portland cement have been studied in the past. Atis reported a decrease in drying shrinkage with the use of fly ash (2003). They created mixtures with varying w / cm (0.28 to 0 34) which they had previously determined to be optimal for maximum compact ability using the vibrating slump test (Cabrera and Atis, 1999). These optimal w/cm were used in creating zero slump concrete mixtures and achieving workability by using a carboxylic type water super-plasticizer. The mixtures were designed containing 100% (control mixture), 50%, and 30% portland cement replacement with a low calcium class F fly ash (ASTM C 618). Two molds of each mixture were made to be tested. The mixtures in the molds were the same except one used a super-plasticizer. They performed unrestrained shrinkage tests on 50 x 50 x 200 nun (2 x 2 x 8 in.) concrete prisms that had been unmolded after 24 hours and then stored at 20 C (68 F) and a relative humidity of 65 percent. Measurements were taken up to six months of age to determine changes in length (drying shrinkage) using a mechanical dial gage. The super-plasticized mixtures containing 0, 50, and 70% fly ash replacement of portland cement exhibited strains equal to 385, 263, and 294 micro strain respectively (2003). When these were compared with the same percent fly ash replacement mixtures but not containing a super-plasticizer they exhibited approximately 50% less shrinkage. The compressive strengths were measured and compared between the control mixture 26

PAGE 46

and the fly ash concrete. The compressive strength of 50% fly ash concrete exceeded the control concrete once it reached 7 days of age. The compressive strength of the 70% fly ash concrete was exceeded by the control at all ages. The 28-day compressive strengths showed a drastic difference. The control, 50% fly ash concrete, and 70% fly ash concrete compressive strengths were 65MPa (9430 psi), 67MPa (9720 psi), and 31 MPa (4500 psi) respectively. Cabrera and Atis' research suggests concrete mixtures with portland cement replacement by approximately 50 percent fly ash and no super-plasticizer to be optimum. Research conducted at the Materials Laboratory at CU-Boulder has shown concrete made with smaller particles of fly ash certainly have some advantages over conventional concrete, but may not be applicable for bridge decks due to its high early strength, high ultimate strength, and low crack resistance (Xi et aI, 2003). Some studies say both Class C and Class F fly ash replacement in concrete increases drying shrinkage and results in increased early cracking with decreased development of tensile strength (Hadidi and Saadeghvaziri, 2005) The research studied in the literature review is tough to decipher; fly ash replacement have mixed results. Its reduction in the rate of stiffness development is helpful in reducing its potential for cracking (Transportation research circular E-CI07, 2006). While the reports are contradictory, the majority of the literature suggests fly ash is beneficial with regards to concrete shrinkage. 27

PAGE 47

3.4.3 (w/cm) The w/cm is the ratio of the weight of the water to the weight of all cementitious materials per cubic yard of concrete (Equation 1). This ratio effects concrete in many ways. The permeability, porosity, ultimate strength and rate of strength gain are all affected by changes in the w / cm. is generally accepted that drying shrinkage increases significantly as the water content increases. ACI 224 Report states that a typical concrete specimen, 134 (225 Ibs/yd3 ) water content, resulted in a drying shrinkage of approximately 300 micro strains. also states that drying shrinkage increases at a rate of30 micro strain per 5 9 Ibs/yd3 ) increase in water content. A study of 12 bridges in Pennsylvania reported crack intensities of 0 to 87m/100m2 (265 with mixture water contents varying from 158 to 173 (267 to 292 Ibs/yd3). An increase in water content showed increased drying shrinkage of approximately 75 micro-strains, indicating that with respect to transverse cracking, mix water content alone was not the significant difference in the performance of bridge decks (Babaei and Purvis, 1995a). Similar articles report concrete with a greater than 0.45 tend to have high porosity and can exhibit substantial drying shrinkage, which results in reduced protection of the reinforcing steel from chlorides (Transportation research circular E-C 1 07, 2006). 28

PAGE 48

3.4.4 Cement Content The cement content has a significant effect on shrinkage and cracking in concrete. Concrete made with higher cement content and a low is more susceptible to cracking than concrete with low cement content and higher (Xi et aI, 2003). Xi et al research and other literature suggest limiting cement content to 470 Ibs/yd3 and that a cement paste volume less than 27.5 % can significantly reduce cracking. However, as high strength concrete has become more common in the industry, it is often encouraged to increase the cement content. Proper measures must be taken for concrete made with increased cement content or it can significantly increase cracking (Transportation research circular E-C 1 07, 2006). Deshpande et al (2007) conducted research using Type II coarse ground portland cement in nine concrete mixtures while varying and aggregate content. It was concluded that a clear trend for shrinkage results from variations in ratio did not exist. At 180 days of age a seemingly sensible pattern of shrinkage occurred in the concrete having the highest aggregate content (80%). The higher the the more shrinkage that occurred; = and 30511: = 0.50. This wasn't the case for the mixtures containing lower aggregate contents of 60 and 70%. They reported the greatest shrinkage occurred in the concrete with a equal to 0.40 (the lowest and a 60% (lowest) aggregate content. Shrinkage was lowest in the concrete with a equal to 0.40 and having 29

PAGE 49

the highest aggregate content of 80%. The research is consistent with other literature in stating that for a given w / c ratio, the shrinkage decreases with an increase in aggregate content. The aggregate acts to restrain the concrete against shrinkage. Adversely, for a given aggregate content, the results of this study show changes in the w/cm and using coarse ground cement to have very little affect on shrinkage (Deshpande et aI, 2007). Cement types vary depending upon the use or the project. Different types of cement produce different temperatures as a result of their hydration processes. Some cement is ground finer and others are coarser. Some cement is designed for high early strength, resulting in a high heat of hydration and high thennal stresses. The resulting stresses make concrete more likely to crack. There are also types of cement designed to gain strength more slowly, corresponding to a lower heat of hydration (Type VII, Type II, and Type IV). Concrete made with these types of cement is expected to result in lower thennal tensile stresses and reduced cracking. Burrows (2003) reports that cracking in bridge decks increased in 1973 when the building code increased 28-day compressive strength requirements from 3000 Ibs.lin.2 to 4500 Ibs.lin.2 The increase in the rate of strength gain causes concrete 30

PAGE 50

to become more brittle and likely to crack. Burrows points out that in 1966 Virginia increased its 28-day compressive strength requirements from 3000 Ibs./in.2 to 4000 Ibs./in.2 It was at this time when bridge deck cracking increased from 11 % to 29%. His research brings attention to numerous Colorado area bridge decks built in the 1950' s that are still in great condition but are being demolished to accommodate a necessary widening of Interstate-25. The bridges of the 1950's had unacceptable 28-day compressive strengths by today's code, but have significantly maintained their structural integrity for half a century. As of 1995, Burrows reports bridge deck cracking in the United States to have increased to 52% of all bridges. This clearly illustrates the upward trend of bridge deck cracking as the required strengths and rate of strength gain continue to increase (Burrows, 2003). Xi et al suggest using Type II cement and avoiding finely ground cement and/or Type III cement (2003). Cements with high alkali content, high C3S and C3A contents, low C4AF, and high fineness have an increased development of strength and are therefore more likely to crack. This is another reason the research circular raises caution in using Type III cement for bridge decks (Transportation research circular E-C 1 07, 2006) 31

PAGE 51

3.4.5.2 Coarse-Ground Cement Research conducted by Deshpande et al for the University of Kansas Research Center show significantly reduced shrinkage in concrete using Type II coarse ground cement versus Type VII cement. In addition, Deshpande examined the effect of aggregate content and on shrinkage (2007). A program consisting of three concrete mixtures made with Type 1111 portland cement and three mixtures made with Type II coarse-ground cement. The was 0.40 for all six mixtures but the aggregate content varied between 60, 70, and 80% for each type of portland cement. At 180 days of age, free shrinkage tests measured significantly lower shrinkage strains in the Type II coarse-ground cement having the highest aggregate content (80%) than the shrinkage strain measured in the concrete made with the Type IIII portland cement having the lowest aggregate content (60%). These strains tapered off near 180 days of age and, at 365 days of age, illustrated an insignificant amount of continued shrinkage strain (Deshpande et aI, 2007). This suggests that both Type II coarse ground cement and mixtures with a higher aggregate content are more suitable for use in crack resistant concrete bridge decks. Brewer and Burrows (1951) tested three cement clinkers ground to finenesses ranging from 1200 to 2700 cm.2 (186 to 419 in.2), in 300 cm.2 increments. They performed tests similar to ASTM C 1581 but using an apparatus created before the 32

PAGE 52

standard was adopted by ASTM. They also perfonned unrestrained shrinkage tests on mortar bars. Restrained shrinkage tests showed concrete made with coarse ground cement resisted cracking longer than the more finely-ground cement concrete. When the coarse-ground cement rings cracked, the unrestrained shrinkage mortar bars were examined further. They discovered the coarse-ground cement mortar bars shrank forty percent more and at a slower rate than the more finely-ground cement concrete. In the end the coarse-ground cement concrete shrank as much as forty percent more than the more finely-ground cement bars due to drying shrinkage It was concluded that mortars made with coarse ground cement are significantly more resistant to cracking than more finely-ground cement concrete due to drying shrinkage (Brewer and Burrows, 1951). 3.4.5.3 Shrinkage compensating cements (SCC) are another cement type currently being studied in the United States. Type K cement (ASTM C845-80) creates an amount of expansion when the concrete is hardening, in an effort to counteract autogenous shrinkage and drying shrinkage. ACI 223R-90 (1992) illustrates the specifications concerning the use of expansive cement. According to Xi et aI, the problem with designing concrete using expansive cement is predicting the amount of expansion necessary for each individual project (2003). Krauss and Rogalla perfonned 33

PAGE 53

research using Type K cement and reported two specimens used in restrained ring shrinkage testing didn't experience significant cracking (1996) Surface cracking occurred but no distinct cracks developed. The research shows ring strains decreased to a constant level without cracking. They also examined a SCC containing an ettringite forming additive. Control mixtures cracked at an average of 20 days and the SCC concrete time to cracking extended to around 36 days. Researcher's state the restrained ring shrinkage test has merit but field performance will vary from laboratory results. Perragaux and Brewster investigated several bridge decks for the New York State Department of Transportation in 1992 Results varied as they compared the bridge decks made using shrinkage compensating cement with surrounding structures previously made using Type II cement. In some structures it was believed that SCC reduced shrinkage by 25% and in some cases the SCC structures cracked more than the Type II cement structures. The research concluded shrinkage compensating cement is not advantageous when compared to Type II cement (Perragaux and Brewster, 1992). Studies performed by the Ohio and New York State Departments of Transportation have returned mixed reviews concerning shrinkage compensating cement. Ohio reported success with the cement in bridge decks while New York had issues with durability (Philips et aI, 1997). In 1989 Purvis performed research

PAGE 54

on concrete slabs made with SCC and found the final net drying shrinkage of SCC slabs was less than slabs made with Type I cement, but the SCC slabs experienced more creep. Because shrinkage is mostly a paste property, it makes sense that increasing the aggregate content decreases shrinkage. Aggregates help by providing restraint to shrinkage while occupying space within the concrete matrix; a space that would otherwise be occupied by additional cement paste. This also helps to create a more economical project because cement is the most expensive material used in making concrete (Transportation research circular E-CI07, 2006). However, aggregates themselves may be responsible for shrinkage. The use of highly absorptive aggregates has proven to result in increased shrinkage. They are more compressible and therefore allow for higher shrinkage. Some may shrink an appreciable amount themselves by the time they are completely dry (Transportation research circular E-C 1 07, 2006). Studies performed by Deshpande et at (2007) examined a program consisting of nine concrete mixtures made with Type II, coarse-ground portland cement. The were 0.40,0.45, and 0.50 and the aggregate contents were 60, 70,

PAGE 55

and 80%. Three mixtures were made with a 0.4 w / c and varied the aggr e gate content from 60, 70 and 80% and this was done for all three w /c. is clear that a delicate balance of aggregate content and ratio are necessary to determine the most appropriate concrete design mixture for crack resistant bridge decks. At 180 days of age a trend developed showing s i gnificantly less shrinkage occurring in the two concretes made with the higher aggregate content (80%) and the lower aggregate content (60%) concrete. The 80% aggregate concrete experienced less shrinkage than the 60% aggregate content concrete. The smallest strain was produced by the highest aggre g ate content mixture having the lowest w/c Accordingly, the highest strain (305Jl) produced in the three mixtures was in the concrete made with an aggregate content of80% and a of 0.50 (highest This was not the case with the other six mixtures. The mixtures containing an aggregate content of 60% and 70% did not follow the same trend as the concrete made with an aggregate content equal to 80%. Although the three mixtures with a 70% aggregate content produced significantly lower strain (360than the 60% aggregate content mixtures it wasn t the mixture with the highest w / c producing the largest amount of shrinkage (Deshpande et ai 2007) 36

PAGE 56

3.4.7 Aggregate 3.4.7.1 The type of aggregate used in concrete can also affect shrinkage. Tests have shown quartzite aggregate to exhibit significantly lower shrinkage strains than concrete made using limestone aggregate. When comparing concrete made using granite, limestone, and quartzite aggregate, the shrinkage strain values at 30 days were 283, 320, and 340 micro strain respectively. These results show that granite aggregate allows for the least amount of shrinkage (Deshpande, Darwin, and Browning, 2007). Xi et al also state that the larger the maximum size of the aggregate the smaller the resulting shrinkage. They report that when the cement paste shrinks, it cannot pull the larger surrounding aggregate closer since they are already in close contact. Micro cracks will develop but so long as they don't develop into larger cracks, the concretes ability to resist cracking is considered to be enhanced and shrinkage reduced (2003). 3.4.7.2 Aggregate Similar research conducted by Meyerson, Mokarem, and Weyers for the Virginia Department of Transportation (2003) used three types of aggregate; limestone, gravel, and diabase. Type 1111 portland cement mixtures with no SCMs were 37

PAGE 57

examined in the first three programs Each program consisted of a range of The fourth program of mixtures were designed with cement replacements by 40% (by weight) ground granulated blast furnace slag (grade 120), Class F fly ash, and a pure amphorous micro-silica each conforming to their appropriate standards ASTM C 989-98, ASTM C 618-97, and ASTM C 1240-97, respectively. At 7,28, and 90 days of age compressive strength tests were performed following ASTM C 39-98 and 102 mm x 203 mm (4 x 8 inch) test cylinder specimens were fabricated according to ASTM C 192-98 A trend of compressive strength developed with 100% portland cement mixtures having varying ranging from 0.42 to 0.49. The first program had of 0.49, 0.47 and 0.46 and the second program consisted of of 0.45 0.43, and 0.42 and both incorporated limestone diabase, and gravel mixtures respectively. In Mokarem et aI's research (2003) the gravel aggregate concrete mixtures consistently had the highest compressive strength but in most cases it was not significantly stronger than the diabase aggregate. However, the limestone aggregate mixtures consistently produced significantly lower compressive strengths than both the diabase and gravel mixtures. As expected the compressive strengths correlated to the the highest compressive strength correlating to the concrete having the lowest and the lowest compressive strength correlating to the concrete having the highest These results were then tested with a third program of 100% portland cement mixtures to verify that it was 38

PAGE 58

not the limestone aggregate properties alone that caused a reduced compre s sive strength. The program included mixtures having ratios of 0.33 0 35 and 0.39 for the limestone, gravel and diabase mixtures, respectively. These are the lowest of any of the programs and this is the largest variation ratios examined in Mokarem et aI's research (2003). At 7 28, and 90 days of age the limestone mixtures compressive strength significantly exceeded that of the gravel and diabase mixtures. At 7 days of age, the compressive strengths measured 7150, 6260, and 6070 lbslin ? (503, 440, and 427kg/cm. 2 ) for the limestone gravel and diabase mixtures respectively. At 28 and 90 days of age the compressive strength continued to follow this trend These results illustrate the inverse proportionality between compressive strength and 3.5 Unrestrained Shrinkage Test Mokarem et allater performed standard tests to determine length change according to standard ASTM C 157-98. Recall, the first program of mixtures had w / c of 0.49, 0.47, and 0.46 for limestone, diabase, and gravel mixtures, respectively. The diabase aggregate concrete experienced the greatest percent length change at almost every age, although the percent changes in length between the three aggregate type mixtures were insignificant up to 56-days of age. The following unrestrained shrinkage data references programs one, two and three and the limestone, gravel, 39

PAGE 59

and diabase concrete mixtures, respectively. At 56-days of age, program one percent length changes were -0 0380, -0.0367, and -0.0392, program two percent length changes were -0.0342, -0.0323, and -0.0392, and program three percent length changes were -0.0321, -0.0328, and -0.0364. After 56-days of age, the diabase aggregate mixtures made with only portland cement began to experience significantly greater percent changes in length than both the limestone and gravel aggregate mixtures, while they continued to experience insignificantly different percent changes in length from one another. The results clearly show an increase in rate of length changes at later ages. At 120 days of age program one percent length changes were -0.0431, -0.0432, and -0.0490 program two percent length changes were -0.0401, -0.0384, and -0.0457, and program three percent length changes were -0.0367, -0.0380, and -0.0453. At 180 days of age, program one percent length changes were -0.0468, -0.0462, and -0.0541, program two percent length changes were -0.0442, -0.0419, and -0.0514, and program three percent length changes were -0.0394, -0.0415, ad -0.0494. Recall the second program consisted of mixtures with of 0.45, 0.43, and 0.42 (the middle range of program examined) for the limestone, diabase, and gravel mixtures, respectively. When standard changes in length were measured for this program, the diabase again experienced the greatest percent length change. These tests show something interesting. The gravel and limestone mixtures experienced percent length changes 40

PAGE 60

correlating to their ratio. The limestone concrete having a of 0.45 experienced a greater percent length change than the gravel concrete having a of 0.42. These results support the idea that a higher equates to more water in the mixture and therefore, more shrinkage However, the diabase concrete had a (0.43) in the middle of the three mixtures and yet it experienced a significantly larger percent length change. When the third program having the lowest range of was examined, the unrestrained shrinkage results illustrate a trend correlating the highest (diabase, 0 39) to the largest percent length change, and the lowest (limestone 0.33) to the smallest percent length change. Mokarem et al attribute this to the to the diabase aggregate absorption value of 1.04% versus the limestone and gravel aggregate which had absorption values of 0.48% and 0.75% respectively. These values indicate that the diabase has more voids filled with water than the other aggregate, which can increase drying shrinkage (2003). When comparing the SCM mixtures, researchers examined mixtures containing the same type of diabase aggregate and the same ratio The mixtures containing fly ash experienced the greatest shrinkage. Micro silica and Ground Granulated Blast Furnace Slag (GGBFS) mixtures were insignificantly different from one another. The drying shrinkage in the mixtures containing SCM's exceeded that of the 100% portland cement mixtures being compared against. This is possibly due to the denser concrete matrix created when using 41

PAGE 61

SCM's. Capillary voids are smaller and would exude less water than normal, larger capillary voids according to Mokarem et ai. This is where drying shrinkage primarily occurs (Mokarem, et. aI., 2003) 3.6 In the ASTM C 1581 (AASHTO PP34), Standard Test Method for Determining Age at Cracking and Induced Tensile Stress Characteristics of Mortar and Concrete under Restrained Shrinkage, a concrete ring is cast around a steel ring Before it was adopted as a standard by ASTM, dimensions of both the steel and concrete ring for the test were modified for various reasons. The current standard (AASHTO PP34, ASTM C 1581) specifies the steel ring to have a wall thickness of 0.50 + / 0.05 in. (13 +/0.12 mm), an outside diameter of 13.0 +/0.12 in (330.0 mm), and a height of 6.0 + / 0.25 in. (152.0 +/6.0 mm), machined smooth on all surfaces. The concrete ring molded around the steel ring is 1.50 in. (38.0 mm) thick. The specimens must be transferred to the testing environment within ten minutes of completion of casting. Four strain gauges are mounted at mid-height (offset 90) around the inside of the steel ring. A data logger begins recording strain measurements within two minutes of the rings being placed in the testing environment. As the concrete ring experiences shrinkage (volume decrease), stresses develop resulting from the steel ring restraining the concrete. The time and 42

PAGE 62

micro strain is recorded upon start and micro strain values are recorded by a data acquisition at intervals not to exceed 30 minutes Moist curing of the molds must begin within 5 minutes of the first strain reading Moist curing continues for twenty-four hours using wet burlap, a relative humidity of 50% +/4%, and at a temperature of73 +/-3.50 F. Micro strain averages are recorded at pre-determined days of age and cracking is recorded to the nearest 0.25 day. When cracks occurs the most recently recorded micro strain prior to cracking is examined. This reading is used as a basis for equations which estimate the micro strain at the actual time of cracking. Over time, variations of the ring test have been performed. The dimensions of the rings used for the test were altered several times. Krauss and Rogalla (1996) examined the affect of changing the dimensions of the rings used for the test. They placed shrinkage stresses that were both uniform and stress increasing linearly from the interface between the concrete and the steel, on the steel ring. They expected this to represent circumferential surface drying or drying from either the top or bottom surface. The research discovered that the height of the steel rings affected the shrinkage stresses in the concrete. As the height increased from 76 mm (3 0 in.) to 152 mm (6.0 in.), shrinkage stresses were reduced. Krauss and Rogalla varied the ring thickness from 13.0 mm (0.50 in.) to 25 mm (1.0 in.) but found little difference in the shrinkage stresses or cracking tendency. Thinner steel rings were

PAGE 63

associated with higher stresses in the steel and the stresses in the concrete rings increased as they increased the steel ring thickness (1996). Attiogbe et al examined ring data involving the thickness of the concrete ring versus it's time to cracking. They discovered that the concrete ring thickness was linearly proportional to its time to cracking and that the depth of drying increases proportionally with the square root of drying time (2004). ASTM C 1581 (AASHTO PP34) is regarded by the engineering field to be a valid and extremely valuable standardized test to detennine the durability of concrete, especially when considering concrete cracking in bridge decks. 3.6.1 Restrained Ring Shrinkage Test The restrained ring shrinkage testing (AASHTO PP34-98) was perfonned for a period of 180 days of age, on 42 ring specimens, and strain measurements recorded at 7, 28, 56, 90, 120, 150, and 180 days of age (Mokarem et ai, 2003 ). Average strains were calculated at each of these days and equations based upon the most recent strain record prior to cracking were used to estimate the strain at any other day. In the first program of mixtures, the diabase rings never cracked through the end of the test period. At the end of 180 days, the diabase concrete rings had an average micro strain value of -132f.lc, significantly less than the limestone and gravel concrete rings. The limestone and gravel concrete rings cracked at 125 days 44

PAGE 64

and 117 days, respectively. At cracking, the limestone ring was determined to have an average micro strain value of -23411 at approximately 120 days, meaning it cracked when it reached a value slightly higher than -23411. When the gravel concrete ring cracked it had an estimated micro strain value of -21 011. Mokarem et al report the diabase concrete had a lower modulus of elasticity (MOE) than the limestone and gravel concrete and researchers believe this may have been why the diabase concrete didn't crack. Mokarem et al state that a higher modulus of elasticity concrete is stiffer and possibly able to resist shrinkage in an unrestrained condition, but the stiffer concrete may create higher strains on the ring in a restrained condition. The mixtures from program two didn't crack at all. At 180 days, the average micro strain values for the limestone, gravel, and diabase concrete were -168, -194, and -20011, respectively. Again, the modulus of elasticity is possibly the cause for the trend in micro strain. The concrete associated lowest MOE having the restrained shrinkage strains. The third program had the lowest of the for the limestone, gravel, and diabase concrete mixtures. Only the gravel and diabase experienced cracking at 165 and 172 days, respectively. The diabase and gravel concrete rings both had an estimated micro strain value of -21011 at cracking. Researchers attribute this program's trend in cracking to the Lower should theoretically experience less shrinkage. In the third program the ratio was the lowest for the limestone, which experienced 45

PAGE 65

a significantly lower amount of strain than the diabase and gravel concrete. None of the rings from the fourth program cracked during the 180 day test period. These mixtures contained SCM's and experienced lowest strains of any programs mixtures. Average micro strain values ranged from to for all of the mixtures at 180 days of age. Mokarem et al note that the strain measured for the fly ash concrete was the highest for both the restrained and unrestrained shrinkage tests. The slag concrete measured the lowest average micro strain value at the end of the test period. Researchers looked at data for the only four rings that broke and each ring had a micro strain greater than Therefore, it was estimated that micro strains greater than will result in cracking of restrained drying shrinkage rings. Using data obtained from concrete having an average micro strain value of it was determined that a strong correlation existed. 3.7 Length Change The corresponding length change associated with concrete having restrained shrinkage strain measuring was thought of as a standard. A percent length change that exceeded those resulting from were then said to increase the probability of cracking. Linear equations for each mixture group were used in calculating associated percentage length changes for each mixture group. Percent length changes in excess of -0.0342, -0.0478, and -0.0482 were determined to 46

PAGE 66

correlate with the cracking of the 100% portland cement mixtures in programs one, two, and three respectively. The mixtures containing SCM's would likely crack if percent length changes occur in excess of -0 0516. Mokarem et al concluded that for 100% portland cement mixtures 28-day percent length change should be limited to -0.0300 and -0 0400 at 90 days to reduce the risk of cracking due to drying shrinkage. For SCM concrete, percent length change should be limited to 0.0400 at 28-Days and -0 0500 at 90 days. 3.8 Water reducing admixtures are often used in concrete to increase workability while maintaining a low resulting in higher strength concrete. A lower of a concrete mixture will result in reduced drying and plastic shrinkage. ACI 212 Committee Report (ACI 212 1989) gives detailed information concerning set retarders and set accelerators. Set retarders are sometimes used in bridge deck applications because they offer delayed set times These retarders allow for continuous placement of bridge decks and make the deck less susceptible to cracking due to deflection of the form work during placement. The delayed set time is also accompanied by lower temperatures during hydration which help reduce cracking due to thermal stresses (Transportation research circular E-C 107, 2006). 47

PAGE 67

Xi et al state that there is no definite conclusion on the influence of set accelerators on bridge deck cracking. The use of retarders increases plastic shrinkage, but decreases the heat of hydration and thermal stresses, resulting in decreased plastic shrinkage cracking (2003). Shrinkage reducing admixtures (SRAs) are a new product currently undergoing testing and research. They work by reducing the surface tension of the concrete water which reduces internal stresses thus lowering long-term shrinkage. Concrete the 50% humidity range develop significant capillary stresses which develop into cracks. SRAs reduce these stresses enough to reduce shrinkage cracking and will be incorporated into mixtures used for this research. 48

PAGE 68

Concrete is always going to crack. Even concrete that has been designed accordingly is expected to crack once in service. If concrete cracks during early stages after placement, it immediately begins to degrade the structure. Preventing the early age cracking of concrete is especially important to the CDOT. is the CDOT's responsibility to maintain a safe network of roads, bridges, and highways throughout the State of Colorado. From public safety to keeping an efficient budget, a durable low cracking potential concrete is very effective in accomplishing both of these objectives. A cracked bridge deck not only diminishes the integrity of the structure but jeopardizes the safety of the travelling public. Substantial damage to the structures integrity begins to occur when cracking in the deck surface allows water to penetrate to the reinforcing steel. The resulting corrosion of steel reinforcement shortens the life span of the bridge and increases maintenance costs while the bridge is in service. These factors are unfavorable, specifically to the department of transportation. Winter conditions in Colorado create the need for increased de-icing salt on the road surface to ensure the safety of the traveling public. The increased amounts 49

PAGE 69

of salt (MgCh or magnesium chloride) accelerate the corrosion process when melting snow transports the chlorides through the small cracks to the steel reinforcement. Research has been underway to investigate several factors contributing to the problems surrounding early age cracking in concrete The CDOT currently has specifications for low cracking concrete used for bridge decks; Class H and Class HT concrete. Current specifications require fresh and hardened concrete properties of the concrete to fall within a specific range. While the current Class Hand HT specifications are an improved approach over previously designed bridge deck concrete, the need for enhancement still exists The purpose of this thesis is to design mixtures with material content ranges above and below that of the current specifications It is believed that the current specifications are creating favorable scenarios for early age cracking. The rate of strength gain, magnitude of ultimate strength, permeability, restrained shrinkage strain, and freeze / thaw durability was tested for each of the designed mixtures and their effects on early age cracking examined. Specifically, eleven, low cracking potential, concrete mixtures were designed, batched and tested for this study. Fresh and hardened concrete properties were examined and their individual effect on concrete cracking analyzed 50

PAGE 70

The primary benefit gained from this research is that the CDOT will be in a better position to design and construct crack resistant bridge decks and other concrete structures. Results from this study will provide the necessary infonnation to develop more durable concrete bridge decks. This data will allow COOT to make changes to current specifications for future construction. Ancillary benefits from this research will include a cost savmgs to the CDOT. Developing a crack-resistant concrete will benefit the COOT by providing for longer lasting concrete structures and reducing the annual costs to maintain these pavement structures. 51

PAGE 71

A national survey of state Departments of Transportation was conducted with the objective of obtaining additional information that may aid in the improvement of the current CDOT specification for structural bridge deck concrete. A web-based tool called SurveyMonkey.com (http://www surveymonkey.com/) was used to formulate the questionnaire and analyze the responses. A 38% response rate was obtained for the State DOT survey. Though the response rate was not as high as the study team had hoped, valuable information was gathered from the survey findings. The survey was submitted to state Departments of Transportation (DOT) Materials and bridge engineers. Analysis was performed on the results and aided in the design process of the concrete mixtures created for this research. Responses were received from 19 of the 50 State DOT's, for a 38% return rate. See Figure 5.1. Multiple states DOT's provided more than one response. Most of the two-respondent states included responses from both the Materials and Bridge 52

PAGE 72

Engineer. The survey returned a total of 33 responses; however, only 28 individuals completed the survey. DOT Respondents Map Multiple responses were obtained from six states: Maryland Transportation Authority, Michigan Department of Transportation, Louisiana Department of Transportation, Tennessee Department of Transportation, Nebraska Department of Roads, and the Arkansas State Highway and Transportation Department. A majority of respondents, 95.0%, replied that their state does experience bridge deck cracking. Transverse deck, full width cracking is common and is expected to 53

PAGE 73

occur at early ages in many states. addition the span type (i.e continuous spans) with positive and negative moment regions have affected the frequency of cracking. The Respondents were asked to choose which of the following choices primarily contributes to bridge deck cracking; placement, curing, rate of strength gain, mixture design or the use of admixtures. The majority of responses selected curing to be the primary cause of cracking pertaining to mixture design, placement, rate of strength gain, and use of admixtures ranked from most influential to least influential respectively Settlement and early-age thermal cracking are also mentioned as causes for deck cracking. The Respondents were asked to select at what age their bridge deck concrete typically reaches its ultimate strength; 3 7 14 ,21, 28 or 56 days A majority of states 42.9%, reported achieving ultimate strength at 7 days. Respondents representing 35.7% claim to achieve ultimate strength at 28 days. Of the fourteen responses, no one reported achieving ultimate strength at 3 days of age. The information suggests that it would be beneficial to slow the rate of strength gain for the concrete being designed for this study.

PAGE 74

5.1.5 AASHTO PP34 Ring Test Usage by State DOTs A majority of Respondents, 93.8%, replied that their state does not perform AASHTO PP34. Many agree that shrinkage is an important issue contributing to cracking; however do not perform any shrinkage measuring tests. One response reported using the test, but finding little increased strain and zero cracking. 5.1.6 Mixture Design Issues The respondents had to choose from four choices pertaining to mixture design; water to cementitious material ratio cement content, chemical admixtures, or pozzolans. Half of the respondents report cement content as the major contributor to bridge deck cracking, while 37.5%. report the cause to be the water to cementitious material ratio. Pozzolans were selected only two times and chemical admixtures were not selected by anyone taking the survey. 5.1.7 Mixture Design Modifications Used to Improve Concrete Performance A common adjustment made by many states is the cement content. Reductions in cement content were mentioned; 660lb/yd3 to 611lb / yd3 and 709lb/yd3 to 57 1 Ib/yd3 approach taken by the Minnesota DOT is to reduce the permeability of the concrete with lower paste contents and higher percentages of SCM's. Their latest designs involve straight portland cement. The concern of the Minnesota 55

PAGE 75

DOT is that the use of supplementary cementitious materials (SCM's) result in lower tensile strengths in the first several days resulting in concrete unable to resist restraint cracking. 5.1.8 Shrinkage-Reducing Admixtures Only 21.4% of the responses indicated using shrinkage-reducing admixtures in their states bridge deck concrete. The Michigan DOT abandoned a project involving S R.A.'s claiming it repeatedly "knocked the air out." An ongoing project currently utilizing S.R.A. 's is the Twin Spans Bridge between New Orleans and Slidell Because the project is ongoing LADOT has not yet reported whether it was or was not beneficial. 5.1.9 Shrinkage Compensating Cement Shrinkage compensating cement (Type K, expansive cement) is currently being tested in the United States. Ohio and New York are two of only several states currently utilizing this type of cement. One problem concerning Type K cement is predicting the amount of expansion which will occur. The majority of the respondents 84.6%, reported having never used shrinkage-compensating cement in their bridge decks. 56

PAGE 76

Admixtures may contribute to bridge deck cracking. Seven choices were provided for selection as materials commonly found in bridge deck concrete mixtures. The choices were silica fume, Class C fly ash Class F fly ash, blast-furnace slag water reducing admixtures (super-plasticizers), set retarders, or shrinkage-reducing admixtures. Silica fume was chosen by most respondents as the cause of increased cracking Blast furnace slag and water reducing admixtures were also selected numerous times. Louisiana suspects they are having problems with the compatibility of materials such as cement, admixtures, and fly ash within their mixture. Set retarders and shrinkage-reducing admixtures were chosen least among the provided choices. Contrary to the responses presented in 5.1.10, some responses claim that blast furnace slag and water-reducing admixtures proved beneficial in reducing cracking in bridge decks. Some states also claim silica fume to be beneficial against cracking. Iowa DOT's report lower shrinkage when slag and Class C fly ash are used as a ternary blend. 57

PAGE 77

5.1.12 Water-to-Cementitious Materials Ratio Four ranges ratio were provided for this question; < 0 35, 0.35 < 0.40,0.40 0.45, and 0.45. A majority of respondents, 78.6%, selected 0.40 < < 0.45 as the range for the maximum allowable for their State DOT's concrete bridge deck mixtures. 5.1.13 Curing Practices Curing is mentioned by many respondents to contribute significantly to bridge deck cracking. A significant number of respondents, 81.8% reported changes in their state's curing practices of bridge deck concrete. A common response was that an increase in moist-cure (wet cure) times from 7 to 14 days was beneficial. Another is the application of wet burlap within 30 minutes of placement. The Michigan DOT specifies strict fogging, burlap, soaker hose systems for a continuous 7 day wet cure, but reports that enforcement of these specifications is inconsistent. Also, it was noted that monitoring concrete temperature and protection of the concrete during its early plastic state are essential in minimizing concrete cracking. 5.1.14 DOT Survey Conclusion The results from this survey were utilized when designing concrete mixtures for this study. The information was used by the University of Colorado Denver 58

PAGE 78

research team in conjunction with the CDOT. The survey was successful in finding a solid foundation of infonnation from which to begin designing concrete mixtures. In addition, it should be noted that bridge deck cracking is not an isolated phenomenon in Colorado, rather is experienced in most all states. When designing the survey several options were available for setting up each question. A mistake was made when creating this survey by allowing multiple responses per person per question. The intention was to get a weighted ranking or rating of survey answers. In several questions, a problem occurred when the program attempted to analyze the responses. The answers were lumped into one large category consisting of every response received. As a result, the percentages accompanying any ofthe analysis reports for these questions are inaccurate. In summary, several factors such as cement content and concrete curing were noted as being influential factors resulting in concrete cracking of bridge decks for several DOTs. Reduction in the total cementitious content and 14 day cure times are a few adjustments to the mixture design and curing practices made by State DOTs. Further, many DOTs do not perfonn shrinkage evaluation tests of any kind on their current bridge deck mixtures. 59

PAGE 79

6.1 Design 6.1.1 Literature Review Chapter 6 Experimental Design A primary objective of this research includes providing the Colorado CDOT with an up-to-date investigation into current research involving the same objectives. This involved extensive use of the internet to find applicable information about pertinent previous and current research. The review also included close examination of several published theses from various universities, students, and engineers around the world. This information was used in the design process of the eleven concrete design mixtures tested during this research. 6.1.2 Mixture Design Process Eleven concrete design mixtures were created for research testing. In addition to the DOT survey and literature review, design input was also gathered from meetings with CDOT engineers and other industry professionals interested in the research. 60

PAGE 80

6.1.3 Mixture Designs Ultimately, eleven design mixtures were developed and tested during this research study. See Table 6 .1. Four mixtures were designed to reduce the early age accelerated strength gain by limiting the 7-day compressive strength to 3000 psi. This was accomplished by adjusting the w/cm, cementitious content of the mixture, and percent of pozzolan replacement. In addition, the use of coarse-ground cement was incorporated into several mixture designs. Table 6.1 Mixture Design Matrix Air 7 % % Within the eleven concrete mixture designs are two Class H control mixtures, per current CDOT Structural Concrete Specifications. One mixture 61

PAGE 81

contains the highest allowable percentage replacement of portland cement with fly ash and silica fume (and lowest allowable w / cm) and the other with the lowest allowable percentage replacement of cement with the same (and highest allowable w / cm). All of the mixtures take into account aggregate content, effective replacement percentages of portland cement with supplementary cementitious materials, chemical admixtures, and varying w / cm. An air-entraining agent (AEA) was used to increase durability of the concrete. Air content within these concrete mixtures was expected to coincide with the required percentages per CDOT structural concrete specifications. Cement Type Mixtures #1 (0.38-6.8-FA20-SFS-II), and #3 (0.38-6 8-FA20-SFS-G) are CDOT control mixtures and have identical mixture proportions and w / cm equal to 0.38, but Mixture #3 is made using the Type G, oil-well cement which is more coarsely ground than common Type II cement. Mixture #2 (0.42 / 6.2 / FA16 / SF3.S/II) and Mixture #4 (0.42/6.2 / FA16/SF3.S/G) are the other CDOT control mixtures but Mixture #4 is again made using the Type G, oil-well cement which is more coarsely ground instead of more common Type II cement. 62

PAGE 82

6.1.3.2 Mixture #5 (0.44 / 6.5 / F A30/Il), Mixture #6 (0.44 / 6.5 / F A30 / SF5 / II), and Mixture #7 (0.44 / 6.5/BFS50/Il) have the same w/cm (0.44) but each introduces various amounts of cement replacement with supplementary cementitious materials; 30% Class F fly ash alone, 30% Class F fly ash and 5% silica fume and a mixture containing only 50% blast furnace slag. The 30% replacement of cement with Class F fly ash in Mixture #5 exceeds current allowable CDOT Class H and HT specification replacement percentages of 20%. 6.1.3.3 Mixture #8 (0.44-6.0-FA30-SRA-II) and Mixture #9 (0.44-6 0-FA30-RET-II) are identical in mixture proportions but each incorporates the use of a chemical admixture Both mixtures exceed current allowable CDOT Class Hand HT specification replacement percentages by having a 30% percent replacement of cement with Class F fly ash. Mixture #8 (0.44-6.0-FA30-SRA-Il) utilizes a shrinkage reducing admixture (s.r.a) to help reduce and control the development of shrinkage strain. S.R.A. s are used in the field to help control shrinkage strain development. The s.r.a was a Master BuildersTetraguard and the maximum suggested dosage rate of 1.5gallyd. 3 was incorporated. Chemical properties for the shrinkage reducing admixture are provided in Appendix Mixture #9 (0.44-6.063

PAGE 83

FA30-RET-II) utilizes a set retarder admixture. These admixtures are often used in the field to delay set time when temperatures are high or traffic holds delivery of fresh concrete. The set retarder was a Master BuildersPozzolith 100XR and an average dosage of 3 ounces per one hundred pounds of cementitious materials in the mixture. Chemical properties for the Pozzolith 100XR can be found in Appendix B. Mixture #10 (0.42-6.0-1I-L.W A) is a 100% portland cement mixture made with a substitution of normal weight sand with 250lbs .l yd.3 oflight weight fine aggregate. The aggregate was pre-conditioned (pre-soaked) to a moisture content (m.c.) of approximately 18%. This was an exceptionally high m.c. for aggregate but is done so with the intent of internally curing the concrete The aggregate releases internal water for use in hydration of cement particles over time. Results are expected to be most significant at 56-days of age. Mixture #11 (0.42-6.0-11Norm Wt.) was a control mixture for comparison against the lightweight aggregate concrete mixture. Mixture proportions are identical to Mixture #10 (0.42-6.0-11L.W.A). 64

PAGE 84

Two types of local cement were used in this research study. Holcim Type II portland cement was supplied by Holcim Inc. and used in the fabrication of several concrete mixtures addition coarse-grained cement supplied by GCC Dacotah Cement from Rapid City, South Dakota, was utilized for two mixtures. This type of cement is a Class G, Oil-well cement. Calcium silicate compounds and other calcium compounds containing iron and aluminum make up the majority of this product. was expected that concrete mixtures containing this cement develop strength much slower than mixtures containing the Type II cement promoting less shrinkage and more resistance to cracking. The cement reports supplied by the manufacturers for the Holcim Type II and Dacotah Class G Oil-well cement are included in Appendix However, the cement compounds chemical and physical properties and compressive strength properties for the Class G Oil well cement are shown in Tables 6.2 6.3, and 6.4 respectively. Dacotah Cement Major Compounds: 3CaO.Si02 Tricalcium silicate 2CaO.Si0 2 Dicalcium silicate 3CaO .AI2 0 3 Tricalcium aluminate 4CaO.AI203Fe203 Tetracalcium alumino ferrite CaS04 2H 2 0 Calcium sulfate dehydrate (Gyp s um) 65

PAGE 85

Table 6.3 Class G Oilwell Cement Chemical and Physical Properties Chemical Physical MgO(%) -1.2 S03 (%) --2. 2 -Ignition Loss (%) 0.8 --Equivalent alkalies (%) 0.21 --Insoluble residue (%) 0.29 -C 3 S ---54 C3A ---4 Blaine Fineness 325 Percent Passing No. 325 Mesh, % 84 Free Water, ml Table 6.4 Class G Oilwell Cement Compressive Strength Properties Compressive Strength 8 hours, 100 degree F. at Atm. Press. MPa (psi) 8 hours, degree F. at Atm. Press., MPa (psi) (1613) Pressure Temperature Thickening Time, minutes Thickening Time Test Chemical and physical properties and compreSSIve strength properties for the Holcim Type II cement are shown in Tables 6.5 and 6.6 respectively. 66

PAGE 86

Table 6.5 Holcim Type II Cement Chemical and Physical Properties Chemical Physical MKO(%) --1.2 S03 (%) --3.2 -Ignjtion Loss (%) -2.4 --Equivalent alkalies (%) 0.7 ---Insoluble residue (%) 0.53 -C 3 S --56 C3A ---6 IBlaine Fineness (m2/kg) 396 Table 6.6 Holcim Type II Cement Compressive Strength Properties Compressive Strength 3 Day 28.7 (4170) 7 Day 37.0 (5360) Pressure Temperature Thickening Time, minutes 137 Thickening Time Test 6.2.2 Aggregate Coarse and fine aggregate were obtained from representative sources within Colorado. The UCD Materials Testing Laboratory acquired both the coarse and fine aggregate conforming to the ASTM C33 standard. Bestway Aggregate provided material properties and gradation reports for the aggregate. The aggregate 67

PAGE 87

properties and gradation have been checked and verified to meet Class Hand HT concrete specifications. The coarse aggregate meets the ASTM C33 Size Number 57 and 67 gradation requirements. The coarse aggregate was obtained from a source located in Brighton, CO. The fine aggregate meets the ASTM C33 gradation requirement for concrete fine aggregate. Based upon laboratory tests performed by WesTest of Denver, Colorado, this aggregate has a low potential for deleterious alkali-silica behavior. The material properties data for both coarse and fine aggregate are included in Appendix B. 6.2.3 Chemical admixtures used for water-reducing (workability) and air-entrainment, as well as shrinkage reduction and set time were utilized in the design mixtures for this research. 6.2.3.1 (H.R.W.R.A.) A CDOT approved high range water reducing admixture was incorporated into several of the design mixtures. The admixture was manufactured by W.R. Grace Daracem 19, ASTM C494 Type A and F, and ASTM CIOl7 Type Chemical Properties for the Daracem 19 is provided in Appendix B. 68

PAGE 88

6.2.3.2 Air-Entraining Agent (A.E.A.) A CDOT approved AEA was utilized for the purposes of air-entraining the concrete design mixtures made for this research. The agent was made by W.R.GraceDaravair _A T60 ASTM C 260. Chemical properties for the Daravair A T60 are provided in Appendix B. 6.2.3.3 Shrinkage-Reducing Admixture (S.R.A.) A COOT approved shrinkage-reducing admixture was utilized for the purposes of this research. The admixture was supplied by BASFMaster Builders_Tetraguard_AS20. Tetraguard_AS20 product data sheets are included in Appendix B 6.2.3.4 Set Retarder (RET) A set retarding admixture was utilized for the purposes of this research. The admixture was manufactured by BASFMaster Builders_Pozzolith_lOOXR. Pozzolith _100XR product data sheets are provided in Appendix B. 6.3 Testing The mixtures were tested according to ASTM standards for different characteristics occurring from 1 day of age through 56-days of age and beyond. The batching 69

PAGE 89

followed ASTM C 192 Standard Practice for Making and Curing Concrete Test Specimens in the Laboratory (AASHTO T 126-97 Making and Curing Concrete Test Specimens in the Laboratory). Both fresh and hardened concrete properties were examined for each mixture batched. The fresh concrete properties that were examined include slump (ASTM C 143, AASHTO T 119), unit weight (ASTM C 138, AASHTO T 121), air content (ASTM C 231, AASHTO T 152), and concrete temperature (ASTM C 1064, AASHTO T 309). Hardened concrete properties that were evaluated in this research included compressive strength (ASTM C 39, AASHTO T 22), unrestrained shrinkage testing (ASTM C 157, AASHTO T 160), restrained ring shrinkage testing (ASTM C 1581, AASHTO PP 34), freeze/thaw durability (ASTM C666, Procedure A, AASHTO 161), and rapid chloride ion penetrability (ASTM C 1202, AASHTO T 227). In addition to the durability, strength, and permeability testing of the mixtures, the shrinkage strain within the concrete was the primary focus of this research. Throughout the life of the concrete, shrinkage strain results from internal stresses created from the depletion of water. As concrete ages, water is continuously depleted by both the exposed surface evaporation of water and the continuous hydration of the internal cement particles. Restrained ring shrinkage testing allowed for an investigation into the development of allowable strain/stress 70

PAGE 90

versus time for each mixture before the concrete cracks. A summary table of test procedures is shown in Table 6.7. Table 6.7 Fresh and Hardened Concrete Properties Tests 6.4 Data Analysis Resulting test data collected from this research was compared and used to provide recommendations for modifications to the current Class Hand HT specification, thereby producing a more crack resistant concrete for use as bridge decks by the CDOT.

PAGE 91

7 the summer and fall 2008, the research team experienced difficulty in having the steel rings (per AASHTO PP34) fabricated due to the size ofthe rings. The research team was able to have four rings fabricated in November 2008. The two additional rings allowed for the testing of two mixtures simultaneously. A few weeks after batching the second and third scheduled mixtures an issue occurred with the data acquisition system. Immediately after casting each concrete ring (a total of four) they were moved to the humidity controlled curing room and the strain gages connected to the data logger. The original program written to record strain measurements at thirty-minute intervals and used previously would not begin taking measurements. As a result, a program was written immediately which would record strain measurements at two-minute intervals. After a few hours and several attempts, a makeshift program was completed, successfully compiled and downloaded to the data logger. The two minute interval was chosen

PAGE 92

to accommodate an efficient and timely assurance of a successfully working program. The program completes three cycles (intervals) before it "zeros out" and begins recording accurate strain measurements. This process took two hours using a program with a specified thirty-minute interval and only eight minutes using a program with a specified two-minute program. In the hastiness and confusion of writing a new program at this time, a small detail specifying the size of the strain data table was overlooked. The size of the table recording the data was left at a default record size of 11,060, instead of being changed to an unlimited range. As a result of the program recording strain measurements at two minute intervals the table filled up before the test was completed; stopping recording data before the rings cracked. As a result the second and third scheduled mixtures (0.38-6.8-F A20-SFS-II and 0.42-6.2-F AI6-SF3.S-II, respectively) were re-batched. 7.1.3 Re-Batch of Mixture #1 (O.38-6.8-FA20-SF5-II) and Mixture #2 (0.426.2-FAI6-SF3.5-II) Upon weighing out the aggregate for the second batch of mixtures one and two (0.38-6.8-F A20-SFS-II and 0.42-6.2-F A 16-SF3.S-II) the restock of aggregate used for the CDOT research was on order. There was enough aggregate to satisfy the batch weights, however, the remaining portion of coarse-aggregate contained 73

PAGE 93

noticeably more fines in its make-up than previously observed. The remaining portion of the coarse-aggregate supply was churned to ensure consistency and uniformity before removal from the aggregate bins. In the interest of time and progression of research, the mixtures were batched using the near-end of the University of Colorado at Denver's Materials Testing Laboratory coarse-aggregate supply. The size of these re-batches were reduced to a size which would produce only enough concrete to make two restrained shrinkage rings (per mixture) and enough 4" x 8" cylinders to test compressive strength. The mixture proportions for batch two of each mixture were identical to those used in the first batch. The compressive strength cylinders from each ofthe re-batched mixtures are only being used to compare the mixtures consistency between the first and second batch. The comparison between the mixtures (first and second batches) is being provided to verify mixture consistency since all other specimens fabricated during the first batching of these mixtures (e.g. unrestrained shrinkage beams and freeze/thaw beams) were continued and used to collect research data for these mixtures. The AASHTO PP34 test ran for a few weeks before the error occurred and the rings needed to be re-fabricated. All other test specimens fabricated during the first batch (e.g. unrestrained shrinkage and freeze/thaw) of mixtures one and two (0.386.8-FA20-SFS-II and 0.42-6.2-FA16-SF3.S-II, respectively) did not experience any problems and were used for research data for their respective mixtures. The 74

PAGE 94

restrained shrinkage specimens fabricated during the second batching of the mixtures (0.38-6.8-FA20-SF5-II and 0.42-6.2-FAI6-SF3.5-II, respectively) were used to collect shrinkage strain through test end; cracking. 7.2 Fresh Concrete Properties Fresh concrete tests included temperature, air content, unit weight, and slump. Fresh concrete properties for the eleven design mixtures are listed in Table 7.1. Table 7.1 Fresh concrete properties Mixture Identification Slump Air Content Unit Weight Ambient Temp. Concrete Temp. (in.) (%) (lbs./ft.3) (OF) (OF) 0.38 / 6.8/F A20/SF5/I1 3.0 5.5 142.4 59 58 0.42/6.2/F AI6/SF3.5/I1 4.5 8.0 134.2 56 58 0.38 / 6 8 / F A20/SF5 / 0 3.5 3.4 147 8 59 62 0.42/6.2/F A 16/SF3 .5/0 5.0 9.5 137.2 62 60 0.44/6.5 / F A301II 8.0 4.5 143.8 62 59 0.44/6.5 / F A30 / SF51II 6.5 9.0 135.8 72 69 0.44 / 6.5 / BFS501II 3.5 3 5 146.4 72 68 0.44-6.0-F A30-SRAII 3.0 2.8 147.4 74 71 0.44-6.0F A3 0RET -II 3.0 7.5 141.4 72 71 0.42-6.0-II (L.W.A) 2.5 7.5 138.6 72 72 0.42-6.0-II (Nonnal Wt.) 2.0 7.5 143.0 66 69 7.2.1 Slump Current Class Hand HT specifications do not specify a slump value. For adequate workability the desired slump was 3.5 inches (8.89cm). Although some values fall below the target all eleven design mixtures achieved sufficient 75

PAGE 95

workability to form test specimens. The use of a High Range Water Reducing Admixture (HRWRA) and Air Entraining Admixture (AEA) was required to obtain the needed workability and durability sought for this research. Mixture #1 (0.38-6.8-FA20 SFS-II) vs. Mixture #3 (0.38 / 6.8 / FA20 / SFS / G) and Mixture #2 (0.42-6 2-FAI6-SF3.S-II) vs Mixture #4 (0.42-6.2-FAI6-SF3.S-G) are CDOT Class H control mixtures which examined the effect of coarse-ground cement versus the specified Type II cement. When comparing the slump values between the mixtures made using Type G coarse-ground cement and Type II cement the coarse ground cement concrete mixtures achieved an increased slump average of inch (1.27cm) over the Type II cement concrete mixtures Mixtures #2 (0.42-6.2-FAI6-SF3.S-II) and #4 (0.42-6.2-FAI6-SF3.S-G) have a w / cm equal to 0.42 and required less HRWRA than Mixtures #1 (O.38-6.8-FA20SFS-II) and #3 (O.38/ 6.8 / FA20 /SFS/ G), which both had w / cm equal to 0.38. Mixture #4 (0.42-6.2F A 16-SF3 .S-G) resulted in a slightly higher slump value than those mixtures with a w / cm equal to 0.38.

PAGE 96

Fly ash is known to increase workability. Figure 7.1 shows Mixture #S (0.44/6.5 / F A30 / II) with an increased w / cm of 0.44 and a 30% replacement percentage of cement with fly ash had significantly increased workability. In fact, Mixture #S achieved the largest slump (8.0 in., 20.32 cm). This slump is higher than what is usually desirable in the field. Mixture #6 (0.44 /6.S/ F A30 / SFS / II) is the same mixture but with a S% replacement of cement with silica fume. Silica fume was expected to decrease workability and did so by 1.S inches (3.81 cm) The 50% blast furnace slag mixture decreased workability significantly from the comparison mixtures #S and #6 (Sin. and 3.Sin. respectively. Mixtures #5 A301II), Mixture #6 (0.44/6.S / F A30 / SF5 / II), Mixture #7 Mixture #8 (0.44-6.0F A30-SRAII), and Mixture #9 (0.446.0-FA30-RET-Il) have a w/cm equal to 0.44 and did not require any H.R.W.R.A. for workability. The chemical admixtures used in Mixtures # 8 and #9 did not result in increased workability except due to the added moisture in the concrete. 77

PAGE 97

7.2.1.4 Aggregate Type Mixture #10 (0.42-6.0-II-L.W.A) and Mixture #11 (0.42-6.0-1I-Nonn.Wt.) with a equal to 0.42 and required very little H.R.W.R.A. .. An advantage Mixture #10 (0.42-6.0-II-L.W.A) has over the other mixtures is the use of pre-soaked light weight aggregate (L.W.A.). The additional water in the presoaked aggregate helped to increase slump (0.5in., 1.27 cm). Each of the eleven mixtures attained adequate workability to mold all necessary test samples. Slump test results are shown in Figure 7.1. en .c CJ ..5 ----.... -.---... ---.-.-. .... --... ----------------.-----.. _-, 8.0 +-----------7.5 7 0 +---------6 5 6 0 +----------5 5 5.0 +-------r1I,..-----4.5 +-----4.0 +----3.5 +---.........--3.0 2.5 2.0 1.5 1.0 0 5 0 0 Figure 7.1 Slump Test Results (ASTM C 143, AASHTO T 119) 78

PAGE 98

7.2.2 The use of an AEA was incorporated for all eleven mixtures. Current Class H and HT specifications require air content between 5% 8%. The air content of the research mixtures varied throughout the research. The W.R. Grace air-entraining agent specifies a dosage of 1 fluid ounce per 100 pounds of cementitious materials. This dosage was measured correctly but resulted in random air contents. Previous research using the same dosage rate of AEA has repeatedly proven accurate air content results. The research team believes the error in air content to be caused by excessive cement replacement percentages with cementitious materials (Fly Ash) which caused unforeseen resulting air contents. Although trial batches were made to test the interaction between the various admixtures, the research team believes the interaction between chemical admixtures and high cementitious replacement percentages caused the design mixtures to have variable air contents. Mixture #3 (0.38-6.8-FA20-SF5-G) was batched first and the exact dosage was used for air content designed to be 6.5%. Mixture #3 (3.4%) is lower than the design of 6.5% by a margin of error equal to 48%. As a result, AEA dosages were re-evaluated for more accuracy. Mixture #1 and #2 were batched next. The AEA dosage was adjusted before batching Mixtures #1 and #2. All of the mixtures using H.R.W.R.A. required an amount different from the design to achieve adequate workability. The two mixtures having lower 79

PAGE 99

equal to 0.38 both required more than the design amount ofHRWRA. As a result, the extended mixing time sometimes required to incorporate the H.R.W R.A. uniformly into the mixture essentially deflated the concrete, releasing the entrained air. This is typically the case with the mixtures having lower air contents than 6 5%. Air contents also varied due to experimental replacement percentages of cement with supplementary cementitious materials and the use of chemical admixtures. These experimental mixtures sometimes had unexpected admixture interactions which were not anticipated during design. Air content values are provided in Figure 7 .2. 80

PAGE 100

10.0 9.5 9 0 8.5 8.0 7.5 7.0 6.5 6.0 c 5.5 5.0 c 0 4.5 4 0 .. 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 Figure 7.2 Air Content (ASTM C 231, AASHTO T 152) 7.2.3 Unit Weight The unit weight of each mixture was determined at batching per ASTM C 138. The unit weight is the weight of a unit volume of concrete (Equation 2). 3 ) Equation 2

PAGE 101

The design unit weight was between 138 and 140.5pcffor all mixtures depending upon the amount of supplementary cementitious materials, w/cm, and resulting air content. The unit weight is affected by the air content and a direct relationship can be seen from the data. When the unit weight is greater than the design, the air content is lower than the design, and vice versa. The air content and the unit weight are inversely proportionate. The unit weight of Mixture #1 (O.38-6.8-F A20-SF5-II) is 2 pounds heavier than the design while the air content is 1 % less than design. Less air within the concrete structure translates to heavier materials filling the void spaces (i.e. sand, rock, cement paste). Since the design unit weight for all mixtures was between 138 and 140.5pcf, and 6.5% air content, the same trend can be seen in all mixtures from the data above. Any mixture having air content higher than 6.5% has a unit weight lower than the design of 140.5pcf and vice versa. Again, the air content and the unit weight are inversely proportionate. The various air contents resulted in unit weights both above and below the 6.5% design. A comparison between air content and unit weight is shown for each mixture in Figure 7 .3.

PAGE 102

155 150 145 140 .c 01 135 130 125 Figure 7.3 Unit Weight (ASTM C 138, AASHTO T 121) vs. Air Content (ASTM C 231, AASHTO T 152) 7.2.4 Concrete Temperature The ideal temperature to place concrete is between 50 and 60 degrees Fahrenheit (10 to 16 degrees Celsius), but should not exceed 85 degrees Fahrenheit (29 degrees Celsius) (Mindess et aI, 2003). Excessive temperatures in concrete cause an increase in the evaporation of water from the concrete. This undesirable increased rate of evaporation is the cause of plastic shrinkage and results in 83

PAGE 103

internal, crack-causing stresses. The concrete temperature for the research mixtures ranged from 58 to 72 degrees Fahrenheit (14 to 22 degrees Celsius). None of the mixture temperatures exceeded the recommended maximum temperature. Concrete temperatures are shown in Figure 7.4. Concrete temperatures were assumed to be acceptable for design performance. 80 72 70 60 50 ::I 40 Co 30 20 10 0 Figure 7.4 Concrete Temperature, (ASTM C 1064, AASHTO T 309)

PAGE 104

Hardened concrete tests performed for this study included compressive strength, restrained shrinkage, permeability, and freeze/thaw durability. Compressive strength is an important design aspect of concrete. More importantly, field performance of the designed compressive strength is imperative. Compressive strength was tested for each mixture at 1,3, 7, 28, and 56-days of age. Three cylinders were tested for each mixture on the respective day of age. The compressive strength was found by dividing the compressive load (lbs.) at failure by the surface area of the concrete cylinder tested (in.2 ) (Equation 3). The cylinders were of the dimensions 4in x 8in (1 O.16cm x 20.32cm, radius x diameter). Figure 7.5 depicts a cylinder being tested. Radius of Cylinder: 2in. (5.08cm) Cross-sectional Area: TI x TI(2") !'e = Compressive Strength: 85

PAGE 105

Figure 7.5 Photograph of Compressive Strength Failure (ASTM C 39, AASHTOT22) Mixtures were designed for laboratory research and ideal conditions. According to current CDOT Class Hand HT specifications, the laboratory trial mixture for Class H or HT concrete must produce an average 56-day compressive strength at least 115 percent of the required 56-day field compressive strength (Equation 4) fc 1.15*fc Equation 4 86

PAGE 106

Current CDOT Class Hand HT specifications require a 56-day compressive strength of 4500psi. As a result, the mixtures designed for this research had a design compressive strength of 5175psi. Compressive strengths for all design mixtures are shown in Table 7.2. Table 7.2 Compressive Strength (ASTM C 39, AASHTO T 22) Mixture Mixture AGE Number Identification I-day 3-day 7-day 28-day 56-day lbs.lin? 1 2135 3880 4632 5778 6479 2 A 1216 2644 3182 4161 4643 3 1369 3879 5232 7621 8712 4 0.42/6.2/F A .5/G 601 1437 2266 3472 3931 5 0.44/6.5 / F A301II 974 2575 3422 4764 5467 6 0.44 / 6.5 / F A30/SF5/11 876 1886 2653 3816 4298 7 0.44/6.5/BFS501II 881 3382 5346 6662 6976 8 0.44-6.0-F A30-SRA-II 1392 2932 3496 4817 5685 9 0.44-6.0-F A30-RET-II 1404 3281 3637 4806 5572 10 0.42-6.0-11 (L.W.A) 2844 4347 4754 5807 6273 11 0.42-6.0-11 (Normal Wt.) 2935 4746 5003 5678 5869 7.4.1.1 Mixtures Having Inadequate 56-Day Strength Current CDOT Class Hand HT specifications require a compressive strength of 4500 psi at 56-days of age. In practice these strengths are sometimes achieved as early as 7 days of age. Other state DOT's require only 3500psi at 56-days of age and feel this is adequate strength for bridge decks. Figure 7.6 shows the 56-day 87

PAGE 107

compressive strength results for all mixtures against the current Class Hand HT requirement. Increased air content results in decreased compressive strength. The compressive strength of concrete is reduced by approximately 5% for each 1 % increase in air content (Mindess and Young, 2003). By having an increase of3.0 and 2.5% air the compressive strength of the mixtures would decrease by 15 and 12.5% respectively. Mixture #4 had a 56-day compressive strength of3931 psi, of which 15% is 590psi, totaling 4521 psi. Mixture #6 had a 56-day compressive strength of 4298psi, of which 12.5% is 452psi, totaling 4973psi. This process is referred to as normalizing data. The normalization of compressive strength for air content shows a sufficient strength for these two design mixtures when air content is accurately incorporated into the mixture. Compressive strengths normalized for air content are discussed further section 7.4.2.2. 88

PAGE 108

9000 8500 8000 7500 7000 @. 6500 6000 5500 Cl c 5000 4500 4000 41 3500 3000 41 2500 2000 0 1500 (J 1000 500 0 Figure 7.6 56-Day Compressive Strength (ASTM C 39, AASHTO T 22) Two of the design mixtures did not satisfy the 56-day compressive strength requirement. Both had air contents in excess of the design by 3.0% and 2.5%, or 9 5% and 9.0% respectively. 7.4.1.2 Normalization of Compressive Strength The air content for the mixtures varied from the design of 6,5%. Various air contents resulted from the use of chemical admixtures, supplementary cementitious materials contents, and the resulting mixing times necessary to achieve adequate 89

PAGE 109

workability of the mixture. As mentioned previously, the compressive strength and the air content are inversely proportionate; as air content increases compressive strength decreases. In fact, the compressive strength of concrete is decreased 5% for each 1 % increase in air content (Mindess and Young, 2003). Normalized, 56day compressive strength results accounting for either a higher or lower air content from the design are shown in Table 7.3 and Figure 7.7. Mixture Mixture Air Deign Air AGE Number Identification Content Content I-day 3-day 7-day 28-day 56-day (%) (%) Ibs.lin.1 0.38/6.8/F A20/SF5/I1 5.5 6.5 2028 3686 4401 5489 6155 2 0.42/6.2/F AI6/SF3.5/I1 8.0 6.5 1307 2842 3420 4473 4991 3 0.38/6.8/F A20/SF5/G 3.4 6.5 1157 3278 4421 6440 7362 4 0.42/6.2/FAI6/SF3.5/G 9.5 6.5 691 1653 2606 3993 4521 5 0.44/6.5/F A301II 4.5 6.5 876 2318 3080 4288 4920 6 0.44/6.5/F A30/SF51II 9.0 6.5 986 2121 2985 4293 4835 7 0.44/6.5/BFS501II 3.5 6.5 748 2874 4544 5663 5930 8 0.44-6.0-F A30-SRA-II 2.8 6.5 1135 2389 2849 3926 4633 9 0.44-6.0-FA30-RET-II 7.5 6.5 1474 3445 3819 5047 5851 10 0.42-6.0-11 (L.W.A) 7.5 6.5 2986 4564 4992 6097 6587 11 0.42-6.0-11 (Normal Wt.) 7.5 6.5 3082 4983 5254 5962 6162 90

PAGE 110

: g a. c W WI a. o 9000, ...... .... .. ........ .... .......... ... ................................................ ....... 8500 +-----------8000 +-----------7500 +-----------7000 +-----------6500 6000 5500 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 o Figure 7.7 56-Day Compressive Strength vs. 56-Day Compressive Strength (Normalized for Air Content), (ASTM C 39, AASHTO T 22) When normalized for air content, all eleven mixtures achieved the current CDOT Class Hand HT field specification requiring 4500psi at 56-days of age. However, the design compressive strength was 5175psi, which several mixtures did not achieve.

PAGE 111

of Mixture The restrained shrinkage specimens fabricated during the second batching of mixtures one and two (O.38-6.8-FA20-SF5-II and 0.42-6.2-FAI6-SF3.5-II, respectively) were used to conduct restrained ring shrinkage tests. The other test specimens (freeze/thaw, penneability, compressive strength) were fabricated during the first batch of mixtures # 1 and #2. Since specimens for the same mixture were been fabricated at two different batch times, a comparison of compressive strength has been perfonned for each mixture, batch one and two. Early-age compressive strength results are shown for Mixture #1 and Mixture #3 in Figures 7.8 and 7.9, respectively.

PAGE 112

_. 1i 6000 5500 5000 4500 rn 4000 3500 3000 2500 o 2000 1500 1000 / --..... 0123456789101112131415161718192021222324252627282930 93

PAGE 113

9000 ...... ............ -.-...................................................................... -... .......................... ......... ................. -..... ..... -..... ,.---------, 6000+_-------------------------------------__ QI -_ Q1> .. -o o 1000 / o 1 2 3 4 5 6 7 8 9101112131415161718192021222324252627282930 Figure 7.9 28-Day Compressive Strength, CDOT Control Mixture #2 (0.426.2-FAl6-SF3.5-II), Batch One vs. Batch Two (ASTM C 39, AASHTO T 22) When making the second batch of mixture 1, the coarse-aggregate supply was nearly diminished and contained noticeably more fines in its composite. Mixture 1 (0.38-6.8-FA20-SF5-II), batch two, demonstrated an increase of less than 1 % compressive strength at I-day of age over batch I (2165 vs. 2161 psi). At 3-days of age the compressive strength of batch two had increased 20% over batch 1 (4831 vs. 3880 psi) and 18% at 7 days of age (5680 vs. 4632 psi). This trend continued as the compressive strength at 28-days of age was 20% higher for batch two than 94

PAGE 114

batch one (7234 vs. 5778 psi). It should be noted that at 7days of age, mixture one, batch two achieved within 2% of the compressive strength as batch one achieved at 28-days of age. Mixture two (0.42-6.2-FAI6-SF3.5-II), batch two was batched immediately following the re-batching of Mixture one (O.38-6.8-FA20-SF5-II), batch 2. The coarse-aggregate supply was almost diminished and it again contained noticeably more fines in its composite. It contained an even slightly higher amount of fines than the re-batch for mixture one (O.38-6.8-FA20-SF5-II). The remaining coarse aggregate supply was churned to ensure consistency and uniformity of the last of the rock. There was more than enough coarse-aggregate to satisfy batch weights so the concrete was made and test specimens fabricated. The results show an increased compressive strength between batches one and two of both mixtures. Mixture two (0.42-6.2-FAI6-SF3.5-II), batch two, achieved 33% increased compressive strength at I-day of age than batch one (1812 psi vs. 1216 psi) and 35% by 3-days of age (4086psi vs. 2644 psi). By 7 and 28-days of age the second batch had achieved 32 and 31 % more compressive strength than batch one (4684 vs. 3182psi and 5998 vs. 4161 psi. respectively). should be noted that mixture two, batch two achieved the same compressive strength at 3-days of age as batch one at 28-days of age. 95

PAGE 115

The increased amount of fines in the coarse-aggregate is believed to be the cause for the increased compressive strength observed between batches one and two. The fines act as a source of strength in concrete and the increased amount of fines would have replaced a portion of the larger aggregate. This results in a more dense concrete structure with an increased compressive strength. The second batch of mixtures 1 and 2 were made to cast new restrained shrinkage rings because the data logger stopped recording strain after an insufficient period of time. The test specimens for permeability, freeze / thaw durability and strength were originally made during batch one of mixtures 1 and 2. The batch two cylinder specimens were made to show a similarity in compressive strength so that data from two different batch times (of the same mixture) would be accepted for the purposes of this study. However the increased fines the coarse aggregate created strengths beyond that of batch one of mixtures 1 and 2, creating errors in the comparison. should be noted that both batches one and two for each mixture remade were identical in batch quantities. Compressive strength varied from mixture to mixture at respective days of testing. Supplementary cementitious materials, the type of cement, and the use of chemical admixtures affect the rate of strength gain at both early and late stages of concrete 96

PAGE 116

age. A comparison of the early-age compressive strength and rate of strength gain is of interest when researching shrinkage strain. An increased rate of strength development will result in increased concrete stresses, often leading to cracking (Xi et aI, 200 1). Compressive strength results and the development of strength will also be discussed in the section analyzing shrinkage strain data for the purposes of this study. A comparison of early-age strength gain for all mixtures through 7 days of age is shown in figure 7.l0. 6000 5500 5000 4500 g. 4000 .c C) 3500 c 3000 2500 2000 (/I 1500 c. 1000 0 500 0 Figure 7.10 Early-Age Compressive Strength (ASTM C 39, AASHTO T 22) 97

PAGE 117

Cement Type Mixtures #1 (O.38-6.8-FA20-SF5-II) and Mixture #3 (O.38-6.8-FA20-SF5-G) are identical in batch quantities but they are made using different cement; common Type II cement and Class G oil well, coarse-ground cement, respectively. Coarse ground cement is expected to decrease the rate of strength gain. Also accompanied by a lower heat of hydration, the concrete is expected to develop lower thermal stresses at early ages and therefore, be less susceptible to cracking. It is expected that the Type G cement will gain early-age strength at a rate slower than the common Type II mixture. Data illustrating the early-age strength gain of the CDOT Class Hand HT mixtures made using the typical Type II cement versus the coarse ground cement are plotted in Figures 7.11 and 7.12, respectively. 98

PAGE 118

! a. c QI o // // o 2 3 _0 5t 4 5 6 7 8 99

PAGE 119

':g c 8 o -' / / /'" o 2 3 -+_----cJ -4 5 6 Figure 7.12 Early-Age Compressive Strength, CDOT Control Mixture #2 (0.42-6.2-FA16-SF3.5-II) (Type II Cement) and Mixture #4 (0.426.2-FA16-SF3.5-G) (Type G, Coarse-Ground Cement) (ASTM C 39, AASHTO T 22) 8 As expected at I-day of age, Mixture #1 (0.38-6.8-FA20-SF5-II) made with Type II cement gained 36% more compressive strength than Mixture #3 (0.38-6.8-F A20SF5-G) made using coarse-ground cement, 2135 vs. 1369psi. At the same age, Mixture #4 (0.42-6.2-F A 16-SF3.5-G), developed only 51 % of the compressive strength achieved by Mixture #2 (0.42-6.2-FA16-SF3.5-II) made using Type II cement, 601 vs. 1216psi. This data shows that coarse-ground cement gains strength 100

PAGE 120

at a slower rate than Type II cement at I-day of age. Mixtures #2 and #4 have a w/cm equal to 0.42 (higher than Mixtures #1 and #30.38) and as expected, are gaining strength at a slower rate than Mixtures #1 and #3. However, by 3-days of age, Mixture #3 (0.38-6.8-F A20-SF5-G) recovered to gain as much strength as its Type II counterpart (Mixture #1 (0.38-6.8-FA20-SF5II)), and actually surpassed the Type II mixture, 3880 vs. 3879psi. The magnitude of the two mixtures is the same at 3 days of age but their respective percentages of ultimate strength acquired are significantly different, 60% vs. 45% respectively. At three days of age Mixture #3 (0.38-6.8-FA20-SF5-G) has achieved 51 % of its 28-day strength while Mixture #1 (O.38-6.8-F A20-SF5-1I) achieved 33% of its respective 28-day compressive strength. However, the Type G cement concrete mixture achieved an almost identical magnitude of compressive strength at 3-days of age compared to its Type II counterpart (3880 psi vs. 3879.2 psi). The Type G cement proves to better regulate the rate of strength gain at 3-days of age and younger. As expected, the higher w/cm mixtures continue gaining strength at a slightly slower rate. This slower rate of strength gain will also reduce thermal stresses and cracking. By 3 days of age Mixture #2 (0.42-6.2-FAI6-SF3.5-1I) achieved 46% more compressive strength than Mixture #4 (0.42-6.2-F A 16-SF3.5-G), 2644 vs. 1437psi.

PAGE 121

At 7 days of age, the increased compressive strength of Mixture #2 had been reduced to 29%, 3182 vs. 2266psi. At 7 days of age and younger and with an increased the Type G cement hydrates more slowly. However, with of 0.38 and increased fly ash the development exceeds the Type II mixture. At 7-days of age the coarse-ground cement has begun to gain strength at a similar rate to the Type II mixtures. The four mixtures (0.38-6.8-FA20-SF5-II, 0.42-6.2-FA16-SF3.5-II, 0.38-6.8-FA20SF5-G, and 0.42-6.8-FA16-SF3.5-G) have achieved 80%,77,69%, and 65% respectively, oftheir 28-day compressive strength. The coarse ground cement mixtures continue achieving a slightly slower rate of strength gain. Strength development trends continue through 7-days of age. At 7-days of age Mixture #4 (0.42-6.2-FA16-SF3.5-G) has achieved the lowest compressive strength (2266psi.). This is due to the increased (0.42) in conjunction with a low percentage replacement of cementitious materials while using Type G cement. At 7-days of age the increased air content in Mixture #6 (0.44-6.5-FA30-SF5-II) continued to reduce its strength gain less than its counterpart (Mixture #5 (0.44-6.5FA30-II)). The silica fume replacement typically increases the strength of concrete 102

PAGE 122

however; the increased air content has super ceded the 5% replacement of cement with silica fume and reduced the compressive strength by 22% (3422 vs. 2653psi.). Mixtures # 5, #6, and #7 all have a relatively low I-day compressive strength Ibs/.in.\ See Figure 7.13. This is due in part to the increased equal to 0.44 for all three mixtures. This increased water will reduce the compressive strength throughout the life of the concrete and is proposed by the research team to help reduce restrained-shrinkage cracking. The only mixture with a lower I-day compressive strength is Mixture #4 (0.42-6.2F A I6-SF3 .5-G). This mixture is made with a equal to 0.42, with a low percentage replacement of cementitious materials, and the coarse-ground, Type G cement. The mixture begins to hydrate the coarse particles more slowly and should be at a lower compressive strength at early-ages. Mixtures #5 (0.44-6.5-FA30-II)and #6 (0.446.5-F A30-SF5-II) are similar except Mixture #6 introduces a 5% replacement of cement with silica fume in addition to the original 30% fly ash replacement. It is the increased air content of9% vs. 4.5% that has reduced the I-day compressive strength of Mixture #6 by 10% of mixture #5 (974psi vs. 876psi).

PAGE 123

6000 5500 5000 4500 4000 3500 I::: 3000 CD 2500 (/) 2000 1500 1000 500 o ,6 t.6' .. ... --... / o 2 3 4 5 6 7 8 At I-day of age Mixture #7 (0.44-6.5-BFS50-II) gained strength within 1% of Mixture #6 (0.44-6.5-FA30-SF5-II). This is the only design mixture utilizing blast furnace slag in this research. At 3-days of age Mixture #6 (0.44-6.5-F A30-SF5-II) gained strength at a slower rate than its counterpart (Mixture #5 (0.44-6.5-FA30-II containing 30% fly ash and no silica fume. The air content reduced the 3-day strength by 27% (1886 vs. 2575psi). 104

PAGE 124

At 3-days of age Mixture #7 (0.44-6.5-BFS50-II) began to develop strength more rapidly than Mixtures #5 and #6. In fact, the blast furnace slag mixture increased the 3-day compressive strength by 24% and 44% beyond that of Mixtures #5 and #6 (3382 psi. vs. 2575psi. or i886psi.). Blast furnace slag proves to reduce early age strength gain but accelerates with hydration beyond i-day of age. The compressive strength between mixtures #6 and #7 is skewed. Air contents for these mixtures varied from 9% for Mixture #6 (0.44-6.5-F A30-SF5-II) and 3.5% for Mixture #7 (0.44-6.5-BFS50-II). As a result, the compressive strength for Mixture #6 is lower than designed and the compressive strength is higher than designed for Mixture #7. The compressive strength results for these mixtures would be closer to one another if the 6.5% air content the mixtures were designed with had been achieved. The air content is believed to have varied slightly due to the fly ash and the blast furnace slag percentage replacements. At 7-days of age Mixture #7 (0.44-6.5-BFS50-II) continued its accelerated strength gain and surpassed Mixtures #5 and #6 by 36% and 50% (5346 vs. 3422 psi and 2653 psi). Blast furnace slag proves to reduce early age strength gain but accelerates hydration beyond i-day of age. 105

PAGE 125

At 1 day of age Mixture #8 (0.44-6.0-F A30-SRA-II) (Shrinkage Reducing Admixture) and Mixture #9 (0.44-6.0-FA30-RET-II) (Set Retarding Admixture) have respective compressive strengths of 1392 and 1402psi. At one day of age the set retarder begins to allow hydration to occur and the rate of strength gain began to increase for Mixture #9 (0.44-6.0-F A30-RET -II. At 3 days of age, the shrinkage reducing mixture achieved only 89% of the set retarder mixture, 2932 vs. 3281 psi. 6000 _-_._._----_ _._ __ ._--_ __ ._--_ __ _-_._-_ _._ ._--_._-_-,..---------...., 5500 +--------------------<, 0 44-6.0-FA30-RET-1I 4500 +------------------------1. 3500 c 3000 CP 2500 2000 8 1500 1000 500 o 0 /' /.' 2 3 .. .... 5 6 7 8 106

PAGE 126

From 3 to 7 days of age the rate of strength gain between the two mixtures is comparable but different in magnitude. The rate of strength for the shrinkage reducing mixture began to increase at 3 days of age. At 7 days of age, Mixture #9 (0.44-6.0-FA30-RET-II) only achieved a 4% increased compressive strength over Mixture #8 (0.44-6.0-FA30-SRA-II). Respective 7 day compressive strengths were 3637 and 3496psi. 7.4.1.4.4 Aggregate Type Mixture #10 .42-6.0-11) made with Light Weight Aggregate (LWA.) and Mixture #11 (0.42-6 .0-11) made with Normal Weight Aggregate (NWA) were compared to investigate the effect of internal curing by incorporating the use of pre-soaked, light weight aggregate (sand). should be noted that concrete made using light weight aggregate is not light weight concrete. The unit weight of the mixture made using light weight aggregate falls within the range of normal weight concrete (138.6Ibs.lft.\ At the time of bat ching, the pre-soaked light weight aggregate had a moisture content of 18%. The increased moisture is expected to effectively help to cure the concrete internally. This internal curing is intended to help reduce restrained shrinkage strain in the concrete as it ages. The L W A sand will release moisture back into the mixture rather than absorbing mixture water during hydration The 107

PAGE 127

LWA is also expected to help the hydration process at ages beyond 7 days (Cusson and Hooegeveen, 2006). As shown in Figure 7.15, at 1 day of age both the normal weight aggregate mixture and the light weight aggregate (sand) mixtures gain strength at a similar rate. Mixture #10 .42-6.0-II) made with LW and Mixture #11 (0.42-6.0-11) made with NW A. achieved compressive strengths within 3% of one another; 2844psi vs. 2935psi, respectively. :g c .2: 8 o ....... /;/ / / o 2 3 -_ .... -..... .... .... 4 5 6 8

PAGE 128

Between 3 and 7 days of age, the nonnal weight mixture began to gain strength at a slightly faster rate than, but still similar to, the light weight aggregate mixture. By 7 days of age, the internally cured Mixture #10 (0.42-6.0-II-LW A.) had only achieved 4754psi when Mixture #11 (0.42-6.0-II-NW A.) reached 5003psi; only a 5% difference. 7.4.1.5 Ultimate Strength (28-Day and 56-Day) The rate of strength gain varies for all mixtures from 1 through 56-days of age. In many of the comparisons, the rate of strength gain for one mixture was increased over another and this changed as the concrete aged. The following sections discuss the compressive strength and the rate of strength gain at 28 and 56-days of age. 7.4.1.5.1 Cement Type At 28 and 56-days of age, respectively, the Type G cement Mixture #3 (0.38-6.8F A20-SF5-G) achieved 32 and 34% more compressive strength than the Type II cement Mixture #1 (0.38-6.8-FA20-SF5-II). See Figure 7.16. This shows that at lower equal to 0.38 the Type G cement can in fact control the rate of eady age strength gain, while it does not jeopardize the ultimate strength of the concrete. The slower rate of strength gain should also result in a lower heat of hydration and, in tum, lower thennal stresses, which can help to decrease early-age cracking in 109

PAGE 129

concrete. However at higher w/cm equal to 0.42, both the early age strength gain and ultimate strength of the concrete is reduced by Type G, coarse-ground cement. See Figure 7.17. At this w/cm, the Type G mixture achieved 17% (4161 vs. 3472 psi.) and 15% (4643 vs. 3931 psi.) less compressive strength at 28 and 56-days than its Type II counterpart, respectively. 9000 ,,----------....,-----.8500 .. I 8000 7500 --0.38/6.8/FA20/SF5/G N:'" 7000 -Jl----------'-. ____ 6000 ____ 4500 3500+--4----------------------------1 o O __ o 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 Figure 7.16 Compressive Strength, CDOT Control Mixture #1 (O.38-6.8-FA20SF5-II) (Type II Cement) and Mixture #3 (O.38-6.8-FA20-SF5-G) (Type G, Coarse-Ground Cement), (ASTM C 39, AASHTO T 22) 110

PAGE 130

, c: o / / ",. -+ ---------------o Figure 7.17 Compressive Strength, CDOT Control Mixture #2 (O.42-6.2-FA16SF3.5-II) (Type II Cement) and Mixture #4 (0.42-6.2-FA16-SF3.5G) (Type G, Coarse-Ground Cement), (ASTM C 39, AASHTO T 22) As mentioned above this reduction in rate of strength gain will result in lower thermal stresses and is expected to help reduce restraineds hrinkage cracking. In addition, the reduction in rate of ultimate strength gain should help reduce restrained-shrinkage cracking beyond early-ages of concrete

PAGE 131

At 28-days of age the silica fume in Mixture #6 (0.44-6.5-FA30-SF5-II) would be expected to increase the rate of strength gain over its counterpart (Mixture #5 (0.446.5-FA30-II)). At 28-days of age the silica fume mixture resulted in lower compressive strengths than its counterpart by 20% (3816 vs. 4764 psi.). This trend accompanies the silica fume mixtures due to the high air content of the mixture previously mentioned. As a result, the accelerated strength gain typically associated with silica fume has been removed. Shown in Figure 7.18, the strength gain through 56-days of age for Mixture #5 (0.44-6.5-FA30-II) and #6 (0.44-6.5F A30-SF5II) are consistent. At 56-days of age, Mixture #6 (0.44-6.5F A30-SF5II) achieved only 79% of the ultimate strength (4298 vs. 5467psi) reached by the mixture made with fly ash and cement alone (mixture #5 (0.44-6.5-FA30-II)). At 28-days of age the 50% blast furnace slag Mixture #7 (0.44-6.5-BFS50-II) continued to surpass Mixtures #5 and #6 by 28% (6662 psi. vs. 4764 psi.) and 43% (6662 psi. vs. 3816 psi.), respectively. At 56-days of age, the 50% blast furnace slag achieved an ultimate compressive strength higher than Mixture #5 (0.44-6.5-F A30-II) and #6 (0.44-6.5FA30-SF5-II) by 22% (6976 vs. 5467psi) and 38% (6976 vs. 4298psi), respecti vel y. 112

PAGE 132

Again, it is clear from the data that blast furnace slag greatly increases the rate of strength gain beyond 7 -days of age and this may contribute to increased restrained-shrinkage cracking at ages beyond 7-days of age. Analysis of AASHTO PP34 test results will help determine the exact role of 50% blast furnace slag replacement in shrinkage cracking. "g g o o ---/ / / / V" j ---_ ..... --! o Figure 7.18 Compressive Strength, Mixture #5 (O.44-6.5-FA30-II), Mixture #6 (O.44-6.5-FA30-SF5-II), and Mixture #7 (O.44-6.5-BFS50-II), (ASTM C 39, AASHTO T 22)

PAGE 133

7.4.1.5.3 Chemical Admixtures Beyond 7 days of age the rate of strength gain is very close between Mixture #8 and #9. See Figure 7.19. The set retarder only retarded hydration during the beginning hours of placement and then the rate of strength gain appears to have recovered. 9000 8500 -t----------------__i 0.44-6.0-FA30-SRA-1I -. 0.44-6 .0-FA30-RET-1I 8000 -t--------------__i N-7000-t----------------------__i 6500-t--------------------------i6000-t------------------------__i 5000 4500 ... 3500 .. / 2500 o 2 4 6 8 1012141618202224262830323436384042444648505254565860 Figure 7.19 Compressive Strength, Mixture #8 (0.44-6.0-FA30-SRA-II) (Shrinkage Reducing Admixture) and Mixture #9 (O.44-6.0-FA30RET-II) (Set Retarding Admixture), (ASTM C 39, AASHTO T 22) At 28-days of age Mixture #8 (0.44-6.0-F A30-SRAII) and Mixture #9 (0.44-6.0-FA30-RET-II) achieved almost identical compressive strengths of4817 114

PAGE 134

vs. 4806psi, respectively. is evident from the figure above that the rate of strength gain for Mixture #8 begins to increase over that of the set retarder Mixture #9 at this time The trend continues between the two mixtures at 56-days of age. Mixtures #8 (0.44-6.0-FA30-SRA-II) surpassed the compressive strength of Mixture #9 (0.446.0-F A30-RET -II) by 2%, 5685 vs. 5572psi. It is clear from the rate of strength gain results that the set retarder only retards the mixture long enough to allow for placement. After the initial hours of placement the rate of strength gain assumes an average rate perhaps reducing the rate of strength from 28 to 56-days of age. The rate of strength gain for the shrinkage reducing admixture initially trailed the set retarder mixture up to 7 days of age, at which time it began to increase. At 1, 3, and 7 days of age the shrinkage reducing mixture was below but within 1, 11, and 4%, respectively. At 28 and 56-days of age the shrinkage reducing mixture surpassed the compressive strength of the set retarder mixture by less than % and %, respectively. 7.4.1.5.4 Aggregate Type Beyond 7 days of age the trend in the rate of strength gain is reversed and the NW A mixture begins to trail the L W A mixture, Figure 7.20. By 28-days of age the L W A 115

PAGE 135

mixture has achieved a 2% higher compressive strength than the normal weight aggregate mixture; 5807 vs. 5678psi, respectively. When the two mixtures reached 56-days of age the internal curing of the light weight aggregate mixture has hydrated the cement particles beyond the normal weight aggregate mixture, reaching a compressive strength of 6273 vs. 5869psi respectively. The continued hydration has resulted in a 6% increase in strength by 56-days of age 1i f/) QI o o ....... -! ..... -o 116

PAGE 136

The more permeable concrete is the more susceptible it is to damage caused by infiltration of contaminated water The permeability test performed for this CDOT research is ASTM C 1202 (AASHTO T 227) or the rapid chloride ion penetrability test (RCIP), and was performed at 28 and 56-days of age for each mixture. Section 3 of ASTM C 1202 summarizes this method as monitoring the amount of electrical current passed through 2-inch (50.8mm) thick slices of 4-inch (l01.6mm) nominal diameter cores or cylinders of concrete for a 6 hour period. Elapsed time of testing for this research was 3 hours and then normalized to calculate the 6 hour data. This can be done because the number of coulombs passed begins to stabilize at 3 hours. The samples were prepared first by wet-saw cutting the top finished surface of a 4" x 8" concrete cylinder specimen. The samples were placed under a dry vacuum (approximately 25 inches (63.5 cm) of mercury) in a desiccator for 3 hours. Water was then introduced to the desiccator and the samples completely submerged. A wet vacuum was pulled for 1 hour before being released. The samples were then left to soak in the desiccator, completely submerged in water for 24 hours.

PAGE 137

The samples were then removed from the water and dried. Silicone was placed around each samples edge to form a seal with a rubber gasket. The cylinder was then placed into the test cell as shown in Figure 7.21. A potential difference of 60-volts (direct-current) is maintained across the ends of the specimen. A sodium chloride solution (N aCr) fills one side of the apparatus and sodium hydroxide solution (NaOH+) on the other, each saturating its respective end of the sample.

PAGE 138

ASTM C 1202 makes a correlation between the total charge passed (coulombs) through the concrete sample and its ability to resist chloride ion penetration. Table 7.4 shows the scale used to designate concretes penneability based upon the coulombs passed. > 4000 High 2,000 4000 Moderate 1 000 2 000 Low 100 1 000 Very Low < 100 Negligible The penneability of concrete develops at various rates and to different magnitudes depending upon the w / cm, cementitious content and quantity and types of SCMs it contains Current COOT Class Hand HT specifications require the 56-day penneability not to exceed 2,000 coulombs, or a chloride ion penetrability rating of low." The results for all eleven mixtures are shown in Table 7 .5. Figure 7.21 is a comparison of 28 and 56-day penneability. 119

PAGE 139

1038 873 Low A 3439 Moderate 1965 A301II 2933 Moderate 1789 2163 Moderate 1387 1272 Low 0.44-6.0-F A30-SRA-II 2329 Moderate 0.44-6.0-FA30-RET 3715 Moderate 0.42-6.0II 2396 0.42-6.0-11 2100 120

PAGE 140

4000 r.!I28-day 3S00 3000 .S6-day 1/1 2S00 1/1 1/1 2000 .Q 0 "3 1S00 0 1000 SOO 0 All eleven design mixtures exceeded current CDOT Class H and HT requirements of 'low' permeability at 56-days of age. This requires fewer than 2,000 coulombs to pass at 56-days of age. Figure 7.23 is a comparison of 56 day permeability. 121

PAGE 141

4000 3500 3000 "C 2500 a.. 2000 .c 0 1500 0 1000 500 0 Mixture #1 (O.38-6.8-F A20-SF5-II), and #3 (O.38-6.8-F A20-SF5-G) have identical mixture proportions and equal to 0.38, but each is made using a different type of cement (Type II vs. Class G, coarse-ground, respectively). At 28-days of age, the Type G cement mixture is more permeable than the Type II mixture, allowing 27% more coulombs to pass during testing (873 vs. 685 coulombs, Figure 7.24. 122

PAGE 142

The rate of hydration of the type G, coarse-ground cement results in a slight change in the development of permeability. By 56-days of age, the Type G cement concrete mixture began to more rapidly hydrate and the mixture was no longer more permeable than the Type II, but less permeable allowing 40% fewer coulombs to pass during testing than the Type II mixture (373 vs. 596 coulombs). At a lower equal to 0.38, the Type G, coarse-ground cement concrete mixtures developed a slightly higher permeability than Type II mixtures at 28-days of age and then lower permeability at 56-days of age. The coarse-ground particles are hydrating more slowly than Type II cement at early ages and more rapidly than Type II cement with increased age. This contrast is evident by the drastic change in the number of coulombs passed by the coarse cement mixture from 28 to 56-days of age. Mixture #3 = 0.38), made using Type G cement, showed a decrease in permeability (coulombs passed) by 30% from 28 to 56-days of age while Mixture #1 (0.38-6.8-FA20-SF5-1I) (same proportions but made Type II cement) only decreased by 13% during the same period of time. This is seen again by comparing the other two mixtures having identical mixture proportions and higher with only cement type as a variable. Mixture #4 = 0.42), made using Type G cement showed a decrease in permeability (coulombs passed) by 43% from 28 to 56-days of age while Mixture #2 (0.42-6.2-F A 16-SF3 .5-II) (same proportions but

PAGE 143

made suing Type II cement) only decreased by 20% during the same period of time. This data shows Type G, coarse-ground cement reduces concrete permeability more rapidly than Type II cement at later ages. 4000 3500 121 3000 2500 a.. 2000 0 "5 8 1500 1000 500 0 28 56 Figure 7.24 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results, CDOT Control Mixture #1 (0.38-6.8-FA20-SF5-II) (Type II Cement) and Mixture #3 (0.38-6.8-FA20-SF5-G) (Type G, Coarse-Ground Cement) (Permeability, ASTM C 1202, AASHTO T 227) Mixtures #2 and #4 represent current CDOT Class Hand HT specifications having the maximum allowable equal to 0.42 and lowest allowable replacement percentage of cementitious materials ; 16% fly ash 3.5% silica fume. 124

PAGE 144

At 28-days of age, the Type G cement mixture is more permeable than the Type II mixture, allowing 70% more coulombs to pass during testing (3439 vs. 1038 coulombs). The slower hydration rate of the coarse-ground Type G cement particles in Mixture (0.42-6.2-FA16-SF3.5-G) coupled with the increased mix water results in a drastic change in the development of permeability. By 56-days of age, the Type G cement mixture is still more permeable than the Type II, allowing 57% more coulombs to pass during testing than the Type II mixture (1965 vs. 835 coulombs) The Type G cement mixture began with a moderate permeability rating and fell to a low permeability rating by 56-days of age. When the was increased from 0.38 to 0.42, Type G cement concrete mixtures show a much higher permeability than Type II, cement concrete mixtures at both 28 and 56-days of age. Mixtures #2 and have the highest per COOT Class Hand Ht specification and the lowest percentage silica fume and fly ash replacement. This combination results in a mixture that is more permeable when compared to Mixture #1 (0.38-6.8-FA20-SF5-II) and #3 (0.38-6.8-FA20-SF5-G) which have the lowes t per the COOT specifications and the highest percentage silica fume and fly ash replacement. A combination of more silica fume and a lower typically result in a less permeable concrete, as seen by the results. Mixture (0.42-6.2-FA16 SF3 5-G) has the highest permeability of all the mixtures batched thus far. This mixture has equal to 0.42 but has the lowest

PAGE 145

allowable percentage cementitious materials replacement allowed per C.D.O.T. specifications. This, in conjunction with the Type G cement used, is believed to have caused this significantly higher permeability. 0.42/6.2/FA 16/SF3 .5/11 0.42/6.2/FA 16/SF3.5/G 3000 1--------4' 31 2500 -I--------JoI' o "5 8 1500 -1--------[/ 1000 500 o 28 56

PAGE 146

7.4.2.2.2 Mixtures #5, #6, and #7 have developed moderate permeability at 28-days of age. This is significantly higher than the first three mixtures. This trend is due to higher ratios equal to 0.42 and 0.44 versus equal to 0.38. The increased mix water results in increased permeability. Mixture #5 (0.44-6.5-FA30-II) and Mixture #6 (0.44-6.5-FA30-SF5-II) have the same and fly ash replacement but Mixture #6 introduces a 5% replacement with silica fume. This explains the decreased permeability (coulombs passed) at 28-days of age; 2163 to 2933, respectively. There is a decrease of 26% due to the 5% silica fume replacement. By 56-days of age the silica fume in Mixture #6 (0.44-6.5-F A30-SF5-II) decreased the concrete permeability by 23% compared to the fly ash Mixture #5 (0.44-6.5-F A30-II), 1387 vs. 1789 (coulombs passed). Mixture #5 showed a 39% decrease in permeability from 28 to 56-days of age, while Mixture #6 showed a similar decrease of 36%. The silica fume hydrated primarily during the first 28-days of age, creating the majority of the permeability. As a result, slightly less water remained for continued hydration of the cement particles beyond 28-Days; slowing down permeability development. The mixture made without silica fume had a more even distribution of water for the hydration of cement particles over time. 127

PAGE 147

Mixture #7 (0.44-6.5-BFS50-II), although designed with an increased equal to 0.44, exhibited 'low' permeability at 28-Days of age due to the 50% replacement of cement with blast furnace slag (1272 coulombs passed). This replacement decreased the concretes permeability at 56-days of age to a rating of 'very low' (991 coulombs passed). This is a 22% decrease in permeability from 28 to 56-days of age and is less than Mixture #5 (0.44-6.5-FA30-II) and Mixture #6 (0.44-6.5F A30-SF5II) made using silica fume and Class F fly ash replacement of cement; 39% and 36% respectively.

PAGE 148

'C 2,000 .Q 0 0 0 Figure 7.26 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results, Mixture #5 (0.44-6.5-FA30-II), Mixture #6 (0.44-6.5-FA30-SF5-II), and Mixture #7 (0.44-6.5-BFS50-II) (Permeability, ASTM C 1202, AASHTO T 227)

PAGE 149

12! 2500 ra Q. 1: 2000 o 1500 1000 500 o Mixture #8 (0.44-6.0-FA30-SRA-II-Shrinkage Reducing Admixture) and Mixture #9 (0.44-6 0-FA30-RET-II-Set Retarding Admixture) were batched at the same time. Both the set retarder mixture and the shrinkage reducing mixture had a equal to 0.44 and developed 'moderate' permeability by 28-days of age. Although they have the same water content and are in the same category at 28-days of age, Mixture #9 (0.44-6.0-FA30-RET-II) developed 37% lower permeability than 130

PAGE 150

Mixture #8 (0.44-6.0-FA30-SRA-II); 3715 vs. 2329 coulombs passed, respectively. The permeability for the set retarder mixture decreased 44% between 28 and 56days of age, while the shrinkage reducing mixture decreased 60%. As previously discussed the set retarder allows for more time before hydration and initial set to occur. The admixture also retards the rate of permeability decrease up to 28 days of age. After 28days of age the permeability ofthe concrete develops at an increased rate. +-------1ettrren1H*ass---= ca 0.. 11 o "3 8 o .2B-Oay Figure 7.28 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results, Mixture #8 (0.44-6.0-FA30-SRA-II) (Shrinkage Reducing Admixture) and Mixture #9 (0.44-6.0-FA30-RET-II) (Set Retarding Admixture) (Permeability, ASTM C 1202, AASHTO T 227)

PAGE 151

til 0 "3 8 28 Mixture #10 (0.42-6.0-II-Light Weight Aggregate) and Mixture #11 (0.42-6.0-11Normal Weight Aggregate) have similar fresh and hardened concrete properties up to 28-days of age. is expected for the L W A mixture to have a higher permeability at 28-days. However, the additional hydration (internal curing) 132

PAGE 152

provided from the L W A is expected to help permeability development at ages beyond 28-days. At 28-days of age Mixture #10 (0.42-6.0-II-Light Weight Aggregate) developed 12% lower permeability than Mixture #11 (0.42-6.0-II-Normal Weight Aggregate); 2396 vs. 2100 coulombs passed. These results classify the two mixtures as having 'moderate permeability at 28-days of age. At 28-days of age, Mixture #10 (0.42-6.0-II-Light Weight Aggregate) and Mixture #11 (0.42-6.0-II Nonnal Weight Aggregate) are each within 15% and 5% of meeting the COOT Class Hand HT specification, respectively. By 56-days of age both mixtures easily exceed the current specification Mixture #10 (0.42-6.0-II-Light Weight Aggregate) and Mixture #11 (0.42-6.0-II Nonnal Weight Aggregate) surpassed the specification by 24 and 26% respectively. The internal curing provided by the pre-soaked light weight aggregate increased the permeability by 3% over the normal weight aggregate mixture. The L W A mixture experienced a significant decrease in penneability between 28 and 56 days due to the continued hydration provided by the additional moisture in the aggregate. 133

PAGE 153

4000,----------------------------------------,---------------. .0.42-6.0-11 (L.W.A) ; Q. 1l 2000 o "5 <3 1500 1000 500 o 28 56 Figure 7.30 28-Day and 56-Day Rapid Chloride Ion Penetrability Test Results, Mixture #10 (0.42-6.0-II-Light Weight Aggregate) and Mixture #11 (0.42-6.0-II-Normal Weight Aggregate) (Permeability, ASTM C 1202, AASHTO T 227) 7.4.3 Durability 7.4.3.1 General The permeability of concrete is associated with its ability to resist or allow water to enter. When water freezes it expands by volume. The more permeable concrete is the more water it will allow to penetrate. When water is allowed to penetrate and freezing temperatures (cycles) occur, the water freezes inside the concrete and 134

PAGE 154

expands against the rigidity of the concrete. The volume expansion of the water creates internal stresses that are damaging to the concrete. Air voids within the concrete structure alleviate the stresses caused by the volume expansion. Depending upon the climate, concrete is designed to contain air voids (air content %) to enhance its durability. In areas where freezing temperatures occur less frequently concrete can be made with relatively low air content. Increased air content will improve the durability of concrete in areas like Colorado, where freezing temperatures occur more often or for longer periods of time. As a result, COOT is interested in researching the durability of any concrete proposed for use in Colorado bridge decks. The ability of concrete to resist freeze/thaw cycles translates to durability. A more durable concrete will better resist the harmful effects caused by freeze/thaw cycles. The freeze / thaw resistance test chosen for this research is the ASTM C 666 Procedure A (AASHTO T 161). Figure 7.31 is a photograph of the University 0 f Colorado at Denver, Material's Testing Laboratory, freeze/thaw chamber. 135

PAGE 155

Freeze/Thaw (ASTM C 666, Two freeze / thaw beams were fabricated for testing of all eleven mixtures. The beams were left in the curing tank until 14 days of age. At 14 days of age the beams were removed from the tank and weighed and their initial resonant frequencies measured per ASTM C 666 prior to being subjected to any freeze / thaw cycles. The beams were then placed in individual metal holding containers in the freeze / thaw chamber. Each container was filled with water to completely submerge the beam and freeze / thaw cycles ensued. 136

PAGE 156

The performance of the specimens during the freeze / thaw testing was determined by measuring each specimen s resonant frequency. Two methods of determining the specimen's resonant frequencies included static and dynamic testing procedures After testing the beams were placed back in the freeze / thaw chamber for approximately 28 additional freeze / thaw cycles. The chamber simulates approximately four, six-hour cycles per day (0 to 40 F or _17 to 4C) producing 28 cycles every 7 days. After 28 cycles the beams were removed weighed, and their resonant frequencies measured again. Figure 7 32 and 7.33are photographs of the E-meter and the static durability test apparatus. 7.32 C 666, 137

PAGE 157

7.33 Testing (ASTM C 666, A transmitter sends a frequency from the mid-section of the beam and a receiver at the left-end of the beam receives the frequency. With exposure to freezing and thawing the beam develops cracks and voids internally when the water expands. The more cracks or voids within the beam the more frequency that is lost in transmission and unable to be received at the end of the beam. As the beam deteriorates and more cracks occur inside the resonant frequency diminishes. 138

PAGE 158

U sing the measured resonant frequency and the corresponding number of freeze / thaw cycles the beam has been exposed to, two calculations are possible. The relative modulus of elasticity and the durability factor are values used to describe the durability of concrete. Results from both test methods are shown in the tables below for all eleven mixtures. Table 7.6 Mixture #1 (O.38-6.8-FA20-SF5-II), freeze/thaw results 0 1992 2029 2031 2066 28 1914 1989 1953 1997 56 1914 1963 1953 1947 84 1895 1984 1953 1963 112 1953 1919 1973 1957 140 1914 1906 1973 1934 168 1973 1970 2012 2026 196 1973 1999 2012 1987 224 1973 1971 1992 2030 252 1934 1984 1953 1990 280 1934 1931 1953 1983 316 1992 2008 1992 2017 139

PAGE 159

Table 7.7 Mixture #2 (0.42-6.2-FA16-SF3.5-II), freeze/thaw results Specimen A Specimen B Dynamic Static Dynamic Static Cycles (Hz) (Hz) (Hz) (Hz) 0 1914 1905 1914 1910 28 1836 1850 1875 1859 56 1823 1850 1836 1855 84 1836 1863 1855 1854 112 1855 1841 1875 1847 140 1823 1835 1856 1842 168 1895 1888 1895 1912 196 1895 1872 1914 1874 224 1895 1869 1875 1908 252 1855 1852 1855 1842 280 1850 1855 1875 1855 316 1914 1923 1895 1910 140

PAGE 160

Table 7.8 Mixture #3 (O.38-6.8-FA20-SF5-G), freeze/thaw results Specimen Specimen A Dynamic Static Dynamic Static Cycles (Hz) (Hz) (Hz) (Hz) 0 2188 2194 2188 2204 28 2090 2111 2090 2099 56 2051 2108 2038 2041 84 1061 2098 990 2030 112 2012 2071 1962 2045 140 1992 2052 1992 2011 168 1973 2050 1986 2006 196 1973 2044 1986 1998 224 1927 2009 1921 1972 252 1934 2003 938 1933 280 1914 1997 1914 1929 308 1953 2029 1901 1968 Table 7.9 Mixture #4 (0.42-6.2-FA16-SF3.5-G), freeze/thaw results Specimen Specimen A Dynamic Static Dynamic Cycles (Hz) (Hz) (Hz) Static (Hz) 0 1934 1945 1934 1948 28 1855 1894 1836 1881 36 1914 1925 1914 1910 78 1895 1909 1875 1886 116 1836 1889 1836 1878 134 1875 1888 1836 1878 162 1875 1890 1855 1877 190 1855 1894 1836 1881 220 1875 1921 1855 1896 253 1816 1861 1797 1839 283 1875 1895 1836 1877 313 1875 1871 1836 1863 141

PAGE 161

Specimen A Specimen B Dynamic Static Dynamic Static Cycles (Hz) (Hz) (Hz) (Hz) 0 2051 2013 2051 2031 28 1973 2002 1914 1952 36 2031 2040 1992 2008 78 2012 2025 1992 2003 116 1953 1988 1934 1981 134 1953 1968 1934 1945 162 1973 1977 1934 1963 190 1934 1947 1875 1898 220 1934 1982 1875 1946 253 1895 1925 1855 1900 283 1914 1912 1875 1895 313 1914 1922 1875 1880 Specimen A Specimen B Dynamic Static Dynamic Static Cycles (Hz) (Hz) (Hz) (Hz) 0 1934 1950 1934 1946 42 1895 1889 1875 1894 78 1836 1857 1836 1865 98 1816 1849 1855 1886 126 1836 1860 1875 1889 154 1758 1788 1816 1850 184 1836 1850 1836 1867 217 1738 1786 1758 1797 247 1816 1830 1836 1843 277 1816 1808 1797 1809 308 1797 1840 1758 1857 142

PAGE 162

Table 7.12 Mixture #7 (O.44-6.5-BFS50-I1), freeze/thaw results Specimen A Specimen B Static Dynamic Static Cycles Dynamic (Hz) (Hz) (Hz) (Hz) 0 2090 2112 2168 2176 42 1992 2012 2070 2097 78 1934 1979 2012 2040 98 1927 1979 1953 2013 126 1875 1910 1953 1985 154 1777 1870 1816 1865 184 1797 1869 1914 1977 1680 1763 1758 1858 247 1797 1845 1914 1929 277 1758 1760 1855 1875 308 1758 1752 1855 1839 Table 7.13 Mixture #8 (0.44-6.0-FA30-SRA-I1), freeze/thaw results Specimen A Specimen B Dynamic Static Dynamic Static Cycles (Hz) (Hz) (Hz) (Hz) 0 1465 1456 1504 1467 30 1313 1297 1341 1323 60 1087 1086 1133 1111 100 957 956 918 945 140

PAGE 163

Specimen A Specimen B Dynamic Static Dynamic Static Cycles 0 1973 1972 1973 1962 30 1973 1960 1957 1950 60 1992 1999 1992 1977 100 1934 1930 1953 1933 140 Specimen A Specimen B Dynamic Static Dynamic Static Cycles 0 2051 2048 2012 2011 40 1973 1960 1953 1947 80 Specimen A Specimen B Dynamic Static Dynamic Static Cycles 0 2109 2100 2129 2123 40 2051 2018 2070 2050 80 144

PAGE 164

The relative dynamic modulus of elasticity (Pc), used to calculate the durability factor, is a ratio of the initial frequency (n) to the frequency when the test is terminated (nl). The test ends after 300 freeze/thaw cycles or when the relative modulus of elasticity of the test specimen has diminished to 60% of the initial modulus (the modulus prior to freeze / thaw exposure). Calculation of the relative modulus of elasticity is performed using Equation 5. where: Pc = relative dynamic modulus of elasticity, after c cycles of freezing and thawing, percent, n fundamental transverse frequency at 0 cycles of freezing and thawing n = fundamental transverse frequency after c cycles of freezing and thawing Note 9 of ASTM C 666 states: This calculation of relative dynamic modulus of elasticity is based on the assumption that the mass and dimensions of the specimen remain constant throughout the test. This assumption is not true in many cases due to disintegration of the specimen. However, if the test is to be used to make

PAGE 165

comparisons between the relative dynamic modulus of different specimens or of different concrete formulations P c as defined is adequate for the purpose. The durability factor (OF) is a ratio of the number of cycles at test termination (N) to the number of cycles when the test is to be terminated (M) and is equal to 300 cycles. This ratio is multiplied by the relative dynamic modulus, Pc (%), at N cycles. Calculation of the durability factor is performed using Equation 6. where: DF = durability factor of the test specimen, P = relative dynamic modulus of elasticity at N cycles %, N = number of cycles at which P reaches the specified minimum value for discontinuing the test or the specified number of cycles at which the exposure is to be terminated whichever is less and M = specified number of cycles at which the exposure is to be terminated 146

PAGE 166

The relative modulus of elasticity for both methods is shown in the following tables. Static Relative Modulus 0 2012 100.00 2048 100.00 28 1934 92.38 1993 94.75 56 1934 92.38 1955 91.17 84 1924 91.45 1974 92.90 112 1963 95.20 1938 89.59 140 1943 93.32 1920 87.93 168 1992 98.07 1998 95.22 196 1992 98.07 1993 94.75 224 1982 97.11 2001 95.46 252 1943 93.32 1987 94.18 280 1943 93.32 1957 91.36 147

PAGE 167

Static Relative Modulus 0 1914 100.0 1908 100.0 28 1855 94.0 1855 94.5 56 1829 91.4 1853 94.3 84 1846 93.0 1859 94.9 112 1865 95.0 1844 93.5 140 1839 92.3 1839 92.9 168 1895 98.0 1900 99.2 196 1904 99.0 1873 96.4 224 1885 97.0 1889 98.0 252 1855 94.0 1847 93.8 280 1855 94.0 1871 96.2

PAGE 168

84 1025 22.0 2064 88.1 112 1987 82.5 2058 87.6 140 1992 82. 9 2032 85.3 1.9 028 85. 1 81.9 2021 84.5 224 77. 3 1991 81.9 252

PAGE 169

78 1885 95.0 1898 95.0 1836 90.2 1884 93.6 28 1943 89.8 1977 95.6 36 2012 96.2 78 95.3 1943 89.8 150

PAGE 170

1948 100.0 .3 1861 91.3 1868 91.9 91.0 84.6 88.9 98 1940 83.0 1914 80.8 1797 71.2 151

PAGE 171

The durability factors have been calculated for the first eight mixtures. Further testing is necessary to determine an accurate factor for the remaining three mixtures. Dynamic Static Average Relative Average Relative Modulus Modulus Cycles Frequency (%) Frequency (%) 0 1973 100.0 1967 100.0 30 1965 99.2 1955 98. 8 60 1992 102.0 1988 102.1 100 1943 97.1 1932 96.4 140 1992 102.0 152

PAGE 172

Dynamic Static Average Relative Average Relative Modulus Modulus Cycles Frequency Frequency 0 2031 100.0 2030 100.0 40 1963 93.4 1954 92.7 80 1958 92.9 Dynamic Static Average Relative Average Relative Modulus Modulus Cycles Frequency (%) Frequency (%) 0 2119 100.0 2112 100.0 40 2061 94.5 2034 92.8 80 2051 93.7 The last three mixtures are currently undergoing testing and their durability analysis will be immediately updated as data is recorded. The durability factor calculation is skewed until each test specimen has been exposed to a sufficient number of cycles to discontinue testing. 153

PAGE 173

The durability factors for the mixtures having undergone sufficient testing are shown in Table 7.28. # % 1 A20 / SF51II 316 101.8 103. 3 102.6 5 5 1.5% 2 A 16/ SF3.5/I1 316 106 3 104 3 105.3 8 0 1.9% 3 308 84.8 79.7 82.3 3.4 6.0% 4 313 96.0 96.1 96.1 9.5 0.1% 5 A30 /II 313 92 2 89 0 90.6 4.5 3 5% 6 308 92.4 86 7 89.6 9.0 6.2% 7 308 72.0 73.9 73.0 3.5 2.6% 8 0.44-6.0-F A30-SRAII 60 11.3 11.2 11.3 2.8 0.9% 9 0.44-6.0-F A30-RET -II N / A 7.5 10 0.42-6.0-11 (L.W.A) 7 5 11 0.42-6.0-11 (Normal Wt.) N / A N / A N / A N / A 7.5 As previously mentioned air content has a direct effect on the durability of concrete however; air content alone does not provide sufficient durability. Theoretically concrete with the highest air content would have the highest durability factor is clear from the data above that air content alone does not control the durability of concrete Concrete with low air content will deteriorate more quickly and have lower durability than a concrete mixture with increased air contents; however supplementary cementitious materials contained in the mixture have a significant impact. 154

PAGE 174

7.4.3.2 Durability Analysis 7.4.3.2.1 Cement Type The air content for Mixture #1 (0.38-6 8-FA20-SF5-II) and #2 (0.42-6.2-FAI6SF3.5-II) are equal to 5.5 and 8.0%, respectively. The durability factors for the two mixtures are relatively close, 102.6 and 105.3, respectively. The increased air content of Mixture # 2 did increase the durability of the concrete slightly. The w/cm for the two mixtures was 0.38 and 0.42. Mixture #1 had the lowest w/cm and highest percentage of cement replacement by fly ash and silica fume allowed per current CDOT Class Hand HT specifications, Mixture #2 had the highest w/cm and lowest allowable replacement percentages. Both mixtures prove to be very durable, maintaining relative moduli well above 60% beyond 300 freeze/thaw cycles.

PAGE 175

------------.-----...... -..... -. ---........................... ----.... ---, 114 0 Content 112 0 110.0 Durability Factor 1 08 0 106 0 ::104.0 ; 102.0 +-100 0 -1-<3 98 0 +-96.0 -1-_ 94 0 -1---o 92. 0 -1-90 0 -1-u.. 88 0 +-86 0 -1-84 0 +--:::s 82 0 a 80 .0+---78.0 +--76 0 -1--74 0 -1--72.0 -1---70. 0+---0 38/6.8/FA20/SFS/II 0.42/6.2/FA 16/SF3.S/II Figure 7.34 Durability Factor and Air Content, CDOT Control Mixture #1 (0.38-6.8-FA20-SF5-II) and CDOT Control Mixture #2 (0.42-6.2FA16-SF3.5-II) Mixture #3 (0.38-6.8F A20-SF5-G) counters Mi x ture # I, but it is made using Type G, coarse ground cement instead of Type II. Although the air content in Mixture #3 is lower than Mixture #1 by 2% it is doubtful to be the sole reason for the difference in durability. Mixture #3 has a durability factor 20% greater than Mixture #1, 102.6 vs. 82.3. This is believed to be the result of the coarse ground cement. Mixture #3 did have a relative modulus below 60 % at two different test times. However, the data is inconsistent with the previous and post data and 156

PAGE 176

believed to be in error. As a result the test was continued and the remaining data showed a consistent trend in durability factors 116.0 ,.-------.-.-----.------.-.-.-------.-... ------,.....-----"""\ 114 0 Content 112 0 110 0 108 0 Durability Factor 106.0 ::104 0 -1----11' ; 102.0 100 0 98 0 96 0-1--_ 94 0 -1---92. 0 90 0 88 0 -1---86 0-1---:a 84 0 :s 82 0 o 80 0 -1----78 0 -1----76. 0 74 0 72.0 -1----70 0 -1---Figure 7.35 Durability Factor and Air Content, CDOT Control Mixture #1 (O.38-6.8-FA20-SF5-II) and Mixture #3 (O.38-6.8-FA20-SF5-G) Mixture #4 A counters Mixture #2 A .5111), but is made using Type 0 coarse ground cement instead of Type II. Mixtures #2 and #4 have air contents equal to 8% and 9.5%, respectively. If air content alone affected durability the 1.5% difference would result in a similar difference of durability factors as seen between Mixture # 1 and

PAGE 177

#3 (2% difference in air content to 20% difference in durability factors). However, Mixture #4 (0.42 / 6.2/FA16/SF3.5 / G) had the higher air content but resulted in lower durability. The w / cm and supplementary cementitious materials replacements for the two mixtures are identical. The durability factor peaks at approximately 9% Any mixtures exceeding such an air content have so much air that the concrete isn't strong enough to resist stresses and weakens the concrete. This is believed to be responsible for the difference in durability between the two mixtures. 116 0,-------------,---------, 114 0 112 0 110 0 106 0 +---1/': ::104 0 +-102.0 +-100.0 -1-(.) 98 0 +-96 .0+-_ 94 0+--92 0 90 0 -1--II. 88 0 +--, 86 0 +-:.c 84 0 +-82 0 +---o 80 .0+-78 .0+-76 0 +--74.0 +--72.0 +-70.0+--0.42/6 .2/FA 16/SF3 .5/11 0.42/6.2/FA 16/SF3 5/G (O.42/6.2/FA16/SF3.S/II) (0.42/6.2/FA16/SF3.S/G) 158

PAGE 178

Mixtures #S A301II), #6 and #7 all have the same (0.44) but each introduces various amounts of cement replacement with other supplementary cementitious materials; 30% Class F fly ash alone, 30% Class F fly ash with an addition of S% silica fume, and a mixture containing only blast furnace slag. Respective air content and durability factors were and 90.6, 9% and 89.6, and and 73.0. Again, it is clear that air content alone does not determine the durability of concrete. Mixture #7 with the lowest air content does in fact have the lowest durability, and 73.0. This mixture contains a replacement of cement with blast furnace slag. The reduced air content of the slag mixture significantly decreased its durability. However, Mixture #6 has the peak air content of the three mixtures (9%) and as a result, had the lowest durability. 159

PAGE 179

116 0,----------------------------,-------, 114 0 +-------------__ ___ ----i, Content 112 0 +-----------------1 110 0 +----------------1 108 0 Durability Factor 106 0 +------------------1 :: 104 0 ; 100 0 +---------------1 o 96.0 94 0 (; 92 0 +------t.I 90 0 88.0 :E 86 0 :c 84 0 82 0 c 80.0 78 0 76 0 74 0 72. 0 70 0 Figure 7.37 Durability Factor and Air Content, Mixture #S (0.44/6.5/FA30/II), Mixture #6 (O.44/6.S/FA30/SFS/II), and Mixture #7 (0.44/6.S/BFS50/II) 7.4.3.2.3 Chemical Admixtures Mixture #8 (0.44-6 0-FA30-SRA-II), made with a shrinkage reducing admixture, had the lowest air content (2.8%) of all eleven mixtures. Thus far Mixture #8 is the only mixture whose relative modulus diminished below 60% before exposure to 300 freeze/thaw cycles. The test specimens for this mixture deteriorated much faster than those previously tested, having a relative modulus below 60% at only 60 freeze / thaw cycles. 160

PAGE 180

Mixtures #9 (0.44-6.0-FA30-RET-II) is currently undergoing freeze / thaw testing and require further exposure to cycles before an accurate durability factor can be calculated. Mixture #10 (0.42-6 0-II-Light Weight Aggregate) and Mixture #11 (0.42-6.0-II Normal Weight Aggregate) are currently undergoing freeze / thaw testing and require further exposure to freeze / thaw cycles before an accurate durability factor can be calculated. As test data becomes available, durability analysis of the remaining three mixtures will be performed and immediately updated. 161

PAGE 181

19 0 +--------------------------------------1 IZI Air Content 18 0 +----------------------------------1 16 0 +-------------------------------------1: Durability Factor g 13.0 +---.-(; o 12.0 +--------r,. 11.0 +-----:: 10.0 +-------o 9 0 +---------:. 8.0 +---7 0 +---6 0 +--5 0 +-----o 4.0 +-____ 3 0+----2 .0+----1 0+--O.O+------.-.J 0.44-6 .0-FA30-SRA-1I Figure 7.38 Durability Factor and Air Content, Mixture #8 (0.44-6.0-FA30SRA-II) 7.4.4 Restrained Shrinkage Strain 7.4.4.1 General The method used for this research to measure restrained shrinkage was the restrained ring shrinkage test (ASTM C 1581, AASHTO PP34). Restrained s hrinkage pertains to Class Hand HT research because reinforced bridge decks are restrained Over time the concrete attempts to shrink (or expand) with temperature

PAGE 182

changes and the reinforcement tries to prevent the shrinkage. The prevention of this shrinkage causes stresses, which translate into strain. Since bridge decks are suspended in the air, without earth for support or temperature absorption, they experience more shrinkage strain than the average reinforced roadway. Figure 7.39 is a photograph of a restrained ring shrinkage specimen Each steel ring was instrumented with 4 strain gages which were mounted on the inside circumference, 90 offset at mid-height. A more detailed description of the 163

PAGE 183

AASHTO PP34 (ASTM C 1581) test, including dimensions of the fabricated steel rings and procedures, is included in Appendix A program was written using the software configured for the data logger being used by the research team. The logger belongs to the UCD Materials Testing Laboratory. The program begins recording strain immediately and continues recording measurements in thirty minute intervals The program must be 'zeroed out' each time a new test is started requires one thirty minute interval to zero, another to take the first measurement, and one more before the measurements begin to stabilize and any external vibration removed. As a result, one or two of the initial strain measurements were sometimes omitted from the data because they were inconsistent. Two restrained shrinkage rings were fabricated for each mixture. The rings were immediately placed in a humidity controlled curing room (40% Relative Humidity) and at a temperature of 73 + / 3 F (23 +/2 C). The dowels securing the concrete ring forms to the supporting form were removed and the rings were immediately covered and cured for 24 hours using wet burlap. The strain gages connected to each ring were then connected to the data logger and the test initialized. At 1 day of age the outer mold of each ring was removed and any sharp corners (approximately 90 top edges) were ground smooth and slightly round with a grinding stone. This was done to eliminate any accumulation of stresses at the edges (corners). Test durations per mixture varied depending upon whether or not 164

PAGE 184

the ring cracked and the rate of strain development. Four rings were continuously utilized allowing the research team to test two mixtures simultaneously. Figure 7.40 is a photograph of the AASHTO PP34 test setup. Current CDOT Class Hand HT specifications require concrete mixtures to not crack before 14 days of age. Tests were typically run for 28 to 30 days, and in

PAGE 185

some cases, 50 plus days. The batching schedule for this research was primarily dictated by the designated amount of time provided for this research. Concrete mixtures were compared on the basis of their individual strain development and magnitude at the time the test was discontinued The rate of strength gain plays an important role in the rate of strain development. Discussion from previous sections analyzing compressive strength and development will be utilized in conjunction with the strain data for each mixture. Accelerated strength development results in a higher heat of hydration, or increased temperatures as cement hydrated during the initial set. Increased temperatures result in increased thennal stresses and increase the likelihood of cracking. Mixture #1 (0.38-6.8-FA20-SF5-II) and Mixture #3 (0.38-6.8-FA20-SF5-G) are identical mixtures but #3 is made using Type G, coarse ground cement. Coarse ground cement was incorporated into this research because it is believed to hydrate more slowly than nonnal Type VII cement. The larger particles are expected to take longer to hydrate and develop strength at a slower rate. The reduced rate of strength gain should result in a lower heat of hydration and reduce thennal stresses. 166

PAGE 186

The reduced stresses should provide reduced strain and in effect a more crack resistant concrete. c u; 0 20 10 0.38/6.8/FA20/SFS/II 0 -10 0 38/6 8/FA20/SF5/G -20 -30 -40 1\.,, ___ ____ ___ ___ ____ ______ -60 -70 -80 -90 -100 -110 -120 -130 o 2 4 6 8 10 12 14 16 1820 22 24 26 28 30 32 34 36 38 40 424446 48 50 52 54 56 58 Figure 7.41 Restrained Shrinkage Strain, CDOT Control Mixture #1 (0.38-6.8FA20-SF5-II) and Mixture #3 (O.38-6.8-FA20-SF5G) (ASTM C 1581, AASHTO PP34) The Type G mixture did in fact develop strength more slowly than the type IIII mixture ; however it gained 26% more compressive strength by 56-days of age, 8712 vs. 6479psi respectively. Mixture #3 (O.38-6.8-FA20-SF5-G) gained more

PAGE 187

strength each day than the Type lIII mixture, but this also corresponds to a smaller percentage of its 56-day strength than the Type 1111 gained each day At 1, 3, and 7 days of age the Type 1111 mixture and the Type G mixture gained the following respective percentages of their 56-day compressive strength; 32.9% vs. 15.7%,59.9% vs. 44.5%, 71.5% vs. 60.5%. The strain measurements do not follow the same trend. At the same days of age, the mixtures gained respective percentages of the ultimate strain of the concrete; 8.1 % vs. -2.3%, 24.3% vs. 36.7%,47.6% vs. 62.6%. By 28-Days of age the coarse cement mixture had only achieved only 85% of its ultimate strength while the Type 1111 mixture had reached 96%. The Type lIII mixture proved to be more crack resistant than the Type G mixture. The test was terminated for Mixture #1 (0.38-6.8-FA20-SF5-II) at 54-days of age, while rings one and two were at an average of 122micro strain and still had not cracked. The mixture experienced a slight decrease in strain at 39 days but it was not a crack. Concerning Mixture #3 (0.38-6.8-F A20-SF5-G), rings 1 and 2 cracked at 16 days of age and an average of 90micro strain.

PAGE 188

7 28 56 Figure 7.42 of 56-Day Strength Achieved at Respective Age, Mixture #1 (O.38-6.8-FA20-SF5-II) and Mixture #3 (O.38-6.8-FA20-SF5-G), (ASTM C 39. AASHTO T 22) 169

PAGE 189

o c .. :5 0 Figure7.43 of Ultimate Strain Achieved, CDOT Control Mixture #1 (0.386.8-FA20-SF5-II) and Mixture #3 (O.38-6.8-FA20-SF5-G), (ASTM C 1581, AASHTO PP34) Mixture #2 (0.42 / 6 2 / FA16 / SF3.5/I1) and Mixture #4 (0.42 / 6 2 / FA16 / SF3.5 / G) are identical mixtures but Mixture is made using Type G coarse ground cement vs. Type 1111 cement. A comparison of s train measurements for each mixture is expected to show reduced strain in concrete made with coarse ground cement. 170

PAGE 190

Again, the coarse ground cement developed compressive strength at a decreased rate than the Type IIII mixture. At 1 and 3 days of age, Mixture #2 (0.42 / 6.2 /FAI6/ SF3.5/II) and Mixture #4 (0.42 / 6.2 / FA16/SF3.5 / G) have developed respective percentages of their ultimate strength; 26.2% vs. 15.3% and 57% s. 36.6%. The trend continues through 7 days of age when the mixtures have again achieved 68.5% vs 57.6% respectively. The Type G, coarse cement clearly reduces the rate of strength gain through 7 days of age.

PAGE 191

A common trend with mixtures having w/cm of 0.42 and greater cementitious materials replacement percentage is a small number of negative strain measurements recorded in the beginning stages of the test. This is assumed to be a swelling of the concrete due to the excess water since it occurs less in mixtures having w/cm equal to 0.38 and more with a 0.44. As expected, the shrinkage strain developed according to the trend of strength development and the coarse ground cement mixture developed strain at a reduced rate. At 1, 3, and 7 days of age Mixtures #2 (0.42 / 6.2 / FA16 / SF3.5/II) and Mixture #4 (0.42 / 6.2 / FA16 / SF3.5 / G) developed the following percentages of their ultimate strain; 10.1 vs. -3.9%, 28.5 vs. 25.8%, and 53.4 vs. 49.5%, respectively. The strain was only slightly reduced by the larger cement particles ofthe Type G mixture. At 28-Days of age the Type II mixture had achieved approximately 100% of its ultimate strain because the strain had leveled off for a couple days before cracking at 29 days of age and an average of 108micro strain. Although the coarse cement mixture developed strain at a reduced rate, the mixture cracked at a smaller magnitude of strain (95 vs. 108micro strain) and in only 24 days.

PAGE 192

Figure 7.45 % of Ultimate Strain Achieved at Respective Age, CDOT Control Mixture #2 (0.42/6.2/FA16/SF3.5/II) and Mixture #4 (O.42/6.2/FA16/SF3.5/G), (ASTM C 1581, AASHTO PP34) As with the previous CDOT Class Hand HT control mixture comparison, the Type G, coarse cement mixture did not prove to be beneficial in producing a more crack resistant concrete Both CDOT Class Hand HT control mixtures, Mixture # I (O.38-6.8-F A20-SF5-II) and Mixture #2 (0.42 / 6 2 / F A 16/SF3.511I), proved to be effective against shrinkage strain in the restrained ring shrinkage test (ASTM C 1581, AASHTO PP34). 173

PAGE 193

c 1/1 0 Figure 7.46 Restrained Shrinkage Strain, CDOT Control Mixture #2 and Mixture #4 (ASTM C 1581, AASHTO PP34) 7.4.4.2.2 Supplementary Cementitious Materials Mixtures #5 A301II) #6 and #7 all have the same (0.44) but each introduces various amounts of cement replacement with other supplementary cementitious materials; 30% Class F fly ash alone, 30% Class F fly ash with an addition of 5% silica fume and a mixture containing only 50% blast furnace slag 174

PAGE 194

All three mixtures have higher w / cm and replacement percentages of Class F fly ash than is currently allowable per COOT Class Hand HT specifications. The incorporation of 50% replacement of cement with ground-granulated blast furnace slag in Mixture #7 is also not allowable per current CDOT specifications. All three mixtures developed shrinkage strain at a very slow rate, measuring 1 day strains still negative; -7.0, -4 .7, and -4.3micro strain, respectively. Negative strain values are common among the initial strain measurements previously recorded when the test is initialized. Mixtures previously tested with w / cm equal or greater than 0.42 have demonstrated this trend Mixture #5 (0.44 / 6.5 / F A30 / II) and #6 (0.44 / 6.5 / F A30 / SF5 / II) both contain an increased (30%) replacement of cement with Class F fly ash. In addition to the fly ash Mixture #6 introduces a 5% replacement of cement with silica fume. This is the highest allowable replacement of cement with silica fume per current CDOT Class Hand HT specification.

PAGE 195

c .. 0 -0 Figure 7.47 Restrained Shrinkage Strain, Mixture #5 Mixture #6 and Mixture #7 (ASTM C 1581, AASHTO PP34) The development of strength gain was similar between Mixture #5 (0.44 / 6.5 / F A301II) and #6 (0.44 / 6.5 / F A30 / SF51II). The air content of the silica fume mixture decreased the magnitude of strength but really didn't affect the rate of strength or strain development. At 3 days of age, both mixtures developed approximately 20% of their ultimate strain. At 7 days of age the silica fume mixture has gained approximately 48% of its ultimate strain and the fly ash only mixture had gained approximately 40% of its own.

PAGE 196

By 28-days of age the mixture made using fly ash replacement alone had gained approximately the same amount of its ultimate strain as the fly ash and silica fume mixture, 94.8 and 94.2%, respectively. The silica fume reduced the magnitude of the ultimate strain for Mixture #6 by 7micro strain, 96 vs. 103micro strain. Mixture #7 developed shrinkage strain at similar rates to Mixture #5 A301II) and #6 at 1, 3, and 7 days of age. However, at 28-Days the blast furnace slag mixture has only achieved approximately 80% of its ultimate strain while mixtures #5 and #7 have almost reached their ultimate strain. The blast furnace slag mixture does result in a higher magnitude of shrinkage strain than the mixtures made using only fly ash replacement and with the addition of silica fume; 113micro strain vs. 103 and 96micro strain, respectively. Mixture #7 achieved a higher ultimate strength than Mixture #5 A301II) and #6 6976psi vs. 5467 and 4298psi, respectively. Mixture #5 A301II) cracked at 28 days and an average of approximately 100micro strain. Mixture #6 cracked at 31 days and an average of approximately 95micro strain. Mixture #7 cracked at 32 days and an average of approximately 90micro strain, although the strain continued to gradually increase to an ultimate strain of 113micro strain when the test was discontinued. 177

PAGE 197

28 56 Figure 7.48 % of Ultimate Strain Achieved at Respective Age, Mixture #5 (O.44/6.S/F A30/I1), Mixture #6 (0.44/6.S/F A30/SFS/I1), and Mixture #7 (O.44/6.S/BFSSO/I1), (ASTM C 1581, AASHTO PP34) 7.4.4.2.3 Chemical Admixtures A comparison of strain measurements was performed on the two mixtures incorporating chemical admixtures. Mixture #8 (0.44-6.0-F A30-SRA-II) incorporated a shrinkage reducing admixture, Master BuildersTetraguard_AS20, and Mixture #9 (0.44-6.0-FA30-RET-II) used a set retarder, Master BuildersPozzolith 100XR.

PAGE 198

The maximum dosage rate ofthe s.r.a was used in Mixture #8 (0.44-6.0-FA30SRA-II), at 1.5gal.lyd.3 or 0.19 gallons per the 3.5ft.3 batch size. This converted to 736.1mL per batch. Cost benefit analysis was not included in the scope of this research, but at the maximum dosage rate it is easy to see how the use of such admixtures could quickly increase a large concrete-project budget. The average dosage rate of the set retarder was used in Mixture #9 (0.44-6.0-FA30-RET-II), at 3 ounces per one hundred pounds of cementitious materials. For the batch having 540lbs/yd.3 of combined cement and fly ash, l60z.lyd.3 or, 473.1mLlyd.3 of the retarder was used. Restrained ring shrinkage test (ASTM C158l, AASHTO PP34) results for Mixture #8 (0.44-6.0-FA30-SRA-II) and Mixture #9 (0.44-6.0-FA30-RET-II) are plotted in Figure 7.50. The shrinkage reducing admixture proved to be very effective against shrinkage strain, achieving an ultimate strain of only 73micro strain at 56-days of age. This was the smallest magnitude of strain achieved by any of the mixtures in the first two to three weeks of testing, and exceptionally at 56-days of age. While the development of strain was decreased significantly, strength development was normal and not decreased, as it achieved approximately 75% of its 28-Day strength at 7 days of age, 3496 of 4817psi, respectively. Figure 7.51 shows the

PAGE 199

development of strength through 56-days of age for Mixture #8 (0.44-6.0-F A30SRA-II) and Mixture #9 (0.44-6.0-FA30-RET-II). c .; en 0 The early age compressive strength developed at a comparable rate to the set retarder mixture. At 7 days of age, Mixture #8 (0.44-6.0-FA30-SRA-II) is only slightly behind in strength development but recovers by 28 and 56-days of age. At

PAGE 200

28-Days of age, Mixture #8 (0.44-6.0-FA30-SRA-II) achieves just a small amount of compressive strength above Mixture #9 (0.44-6.0-FA30-RET-II), less than 1 %, at 4817 vs. 4806psi, respectively. The mixtures achieved 56-day compressive strengths within 2% of one another, 5572 vs. 5685psi respectively, and at the same time the ultimate strain of the s.r.a. mixture was reduced by 42% from the set retarder mixture, 125 vs. 73micro strain respectively. 100.0% 95.0% 90.0% 85.0% 80.0% "C 75.0% > 70.0% :c 65.0% 60.0% .c -55.0% s: 50.0% 45.0% 1\1 40.0% c th 35.0% It) 30.0% 0 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% l1li:. --;/ .. 1/ -0.44-6.0-FA30-SRA-11 0.44-6.0-FA30-RET-1I o 2 4 6 8 1 0 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60

PAGE 201

The test was discontinued at 57 days of age due to time constraints on this research project. At 56-days of age, Mixture #8 (0.44-6.0-FA30-SRA-II) shrinkage rings were performing against the development of shrinkage strain far superior to Mixture #9 (0.44-6.0-FA30-RET-II), and all other mixtures surviving as many days of testing, The rings had not cracked. Mixture #9 (0.44-6.0-FA30-RET-II) cracked at 36 days of age and approximately 128micro strain. The shrinkage reducing admixture proved to be very effective when used at the maximum dosage rate. Development of strength was adequate and shrinkage strain was greatly reduced as a result of the admixture. The air content for the mixture was only 2.8% due to the s.r.a. and, as a result, the mixture exhibited poor freeze/thaw durability. At 7 days of age, the set retarder mixture had achieved 43% of its ultimate strain while the s.r.a. mixture had only reached 21 %,53 vs. 15micro strain respectively. The trend continued at 28-Days of age as the set retarder mixture had achieved 92% of its ultimate strain vs. the s.r.a. reaching only 62% of its own, 115 vs. 45micro strain. The set retarder strain measurements are not exceptionally high in magnitude of micro strain but the rate at which the mixture developed the strain is quite high. Increased development rates of strain often lead to cracking in the field and are not beneficial.

PAGE 202

r;.;J 41 -t-------------------
PAGE 203

releases some of its internal water for use in hydrating cement particles as it ages. Figure 7.53 shows the development of strength through 56-Days of age. Strength development is only slightly decreased beyond 1 day of age. The lightweight aggregate is made of expanded shale and is weaker in shear than normal limestone or quartz aggregate. Results show 28-Day compressive strengths to be comparable as increased hydration past 7 days of age causes the rate of strength gain to recover to within 2% of the normal weight aggregate mixture, 5678 vs. 5807psi for #10 and #11 respectively By 56 days of age the continued internal curing from the light weight aggregate mixture has developed 6% more compressive strength of the normal weight aggregate mixture; 6273 vs. 5879psi. 184

PAGE 204

"C CD CD :2 tJ c( -'= "51 c CD >-ca 0 :,e .. ----1 1 ......... o Figure 7.52 of 56-Day Strength Achieved, Mixture #10 (0.42-6.0-II-L.W.A) and Mixture #11 (0.42-6.0-II-Norm.Wt.), (ASTM C 39, AASHTO 22) Mixture #9 (0.44-6.0 F A30-RET -II) only reduced the development of early age strain slightly at 1 day of age, before accelerating past the s.r.a. mixture at 7 days of age. The ultimate strength was not affected but the development of strain was greatly increased. Restrained ring shrinkage test results are plotted in Figure 7 54 The lightweight aggregate developed strain at a decreased rate than the normal weight aggregate mixture The L W A and normal weight mixtures reached 21 vs. 29 % of their ultimate strain at 3 days of age respectively. By 7 days of age 185

PAGE 205

the L W A mixture reached 47% of its ultimate strain and the normal weight aggregate mixture 57%. By 28-Days of age Mixture #10 (0.42-6.0-II-L.W.A) achieved 8% less shrinkage strain than Mixture #11 (0.42-6.0-II-Norm.Wt.); 110 s. 119micro strain respectively. Mixture #10 (0.42-6.0-II-L.W.A) shrinkage rings cracked at 33 days and an average of approximately 125micro strain. Mixture #11 (0.42-6.0-II-Norm.Wt.) shrinkage rings cracked at 34 days and an average of approximately 134micro strain. 186

PAGE 206

."...------------r--------------------, Ui .......... ,..".__-:::> '0 +---------3 7

PAGE 207

c 0 ... --+Results from the use of pre-conditioned lightweight aggregate are expected to be more significant at 56-days of age. The mixtures reached 56-days of age on November l3, 2009 and respective tests were performed. Data will be updated immediately upon analysis. 188

PAGE 208

Shrinkage is a paste property of concrete. This means that the volume shrinkage that occurs is the paste shrinking and not the fine or coarse aggregate. The water is in the mortar paste (cement, sand, and water) of the concrete and this paste is what attempts to shrink around the coarse aggregate. The aggregate restrains against the shrinkage and causes the concrete to crack. One approach is to minimize the paste content in a concrete mixture and therefore, less paste should equate to less shrinkage. will be beneficial for this research to perform a comparison of paste content and the development of strain. Table 7.29 lists each of the mixture properties including paste content.

PAGE 209

Table 7.29 Design mixture properties Cementitious Air Mixture ID w / cm Content Type of Cement ADMIX. Content Paste Volume (%) 0.38 / 6.8 / F A20/SFSIII 0.38 640 Type II 6.5 28% A 16/ SF3.S/II 0.42 580 Type II 6 5 26% Class G Oil Well Cement (Coarse Grained 0.38 / 6.8 / F A20 /SFS/ G 0 38 640 Cement) 6.5 28% Class G Oil Well Cement (Coarse Grained A 16/SF3.S / G 0.42 S80 Cement) 6.S 26% A301II 0.44 611 Type II 6.S 29% 0.44/6.5 / F A30 / SFSIII 0.44 611 Type II 6.S 29% 0.44 / 6.5 / BFSSO/II 0.44 611 Type II 6.S 28% 0.44 / 6 0 / F A30 / SRAlII 0.44 540 Type II SRA 6.S 2S% A30 / RETIII 0.44 540 Type II RET 6.5 2S% 0.42/6.01II-L.W.A. 0 42 564 Type II 6 5 2S% 0.42 564 Type II 6.5 2S% Paste content has long been recognized as a factor in concrete shrinkage. Moderate paste content was a priority in the designing of concrete mixtures used for this research. An average paste content of 28% was consistent for several of the mixtures. For the benefit of this research some of the mixtures were designed with paste contents slightly higher than what is ideal. This was done to examine the

PAGE 210

effect those cementitious materials on shrinkage. The excess paste provided a clearer result of exactly how these cementitious materials effect shrinkage. The two mixtures having the highest paste content are Mixture 5 (0.44 / 6.5/F A30/II) and Mixture #6 (0.44 / 6.5 / F A30 / SF51II). The two mixtures reached average ultimate strains that are comparable with the other mixtures, 103 and 93micro strain. The two mixtures having the lowest paste content (25%) have reached 56days of age and strain measurements were analyzed. Mixture #8 (0.44 / 6.0/FA30 / SRAlII) and Mixture #9 (0.44/6.0/FA30/RETIII) reached an ultimate strain of (73 vs. 105micro strain). Both are comparable to the 29% paste content mixtures. should be noted that all four mixtures have w / cm equal to 0.44. At 1 3, 7, and 28-Days of age, Mixture #8 (0.44 / 6 0 / FA30 / SRAlII) with 25% paste content has achieved less of its ultimate strain at respective days than any of the other mixtures. However, this is not an accurate representation of 25% paste content. Mixture #8 incorporated a shrinkage reducing admixture which decreased its ultimate strain as well as its strain development at all ages. At 1 day of age the 25% paste content mixture containing set retarder achieved more of its ultimate strain than both of the 29% paste content mixtures. From 3 to 7 days of age Mixture 5 (0.44/6.5/F A301II), Mixture #6 (0.44 / 6.5/F A30 / SF51II) and Mixture #9 (0.44 / 6.0 / F A30 / RETIII) reached

PAGE 211

comparable percentages of their ultimate strain; 19.4, 21.5, and 21.1 % at 3 days, respectively. Mixture #5 has 29% paste content and developed strain at a comparable, but increased rate when compared to 25% paste content mixtures. By 28 days of age, increased paste content only slightly increased shrinkage strain. Mixture 5 (0.44/6.5/F A301lI) and Mixture #6 (0.44/6.5/F A30/SF5/II) having 29% paste content reached 95 and 94% of their ultimate strain, while Mixture #9 (0.44/6.0/F A30 / RET/II) was close behind at 92%. At early ages, increased paste content of 4% only slightly increased development of strain. In fact, the ultimate strain of Mixture #9 (0.44 / 6.0/FA30 / RETIlI) surpassed the 25% paste content mixtures by approximately 10 to 15micro strain. 192

PAGE 213

respectively. Mixture # 2 has an increased w / cm but less cement than Mixture #1, resulting in the decreased 2% paste content. At 1 day of age both mixtures have low strain values but each reaches approximately 10% of their ultimate strain. By 3 and 7 days of age Mixture #2 (0.42 / 6.2 / F A 16 / SF3.S/II) with increased w / cm (0.42 vs. 0.38) and decreased paste content (26 vs. 28%) achieved IS and 11 % more of its ultimate strain than Mixture #1 (0.38 / 6.8 / FA20 / SFS/II) respectively. By 28-Days of age Mixture #2 (0.42 / 6.2 /FAI6/ SF3.S/II) with 26% paste content had achieved 100% of its ultimate strength while Mixture # 1 (O.38/ 6.8 / F A20 / SFSIII) with 28% paste content only reached 96%. Mixture #2 (0.42 / 6.2 / F A 16/ SF3.S/II) achieved higher ultimate strain with 26% paste content than Mixture #1 (O.38/ 6.8/F A20 / SFS/II) with 28% paste content. The increased w / cm is believed to be the reason for the increased strain of 12% (127 vs 112micro strain respectively). Paste content didn't seem to affect shrinkage strain alone. Mixtures with increased w / cm and less paste content achieved higher ultimate strains than those with increased paste content and decreased w / cm.

PAGE 214

:E t: :5 '0 3 7

PAGE 215

This thesis evaluated the current CDOT Class Hand HT concrete mixture specification In addition nine other mixtures were investigated to aid in the development of a more crack resistant concrete specification. In total eleven concrete mixtures were design, batched, and tested for their fresh and hardened concrete performance Specifically the designs differed by type of cement w / cm cement content SCMs and use of chemical admixtures. Compressive strength, permeability freeze-thaw resistance and restrained shrinkage cracking were evaluated and reported in this thesis. A summary of the major findings from this study are reported below Slump values were increased slightly with the use of Type G coarse-ground cement. In addition, an increase in slump was also observed when the percentage of cement replacement with fly ash was increased beyond the current replacement levels. 196

PAGE 216

The air content of the mixtures varied between mixtures. The Type G cement didn't seem to have any effect on air content at of 0.38. However, at of 0.42 the air content greatly increased with the Type G cement concrete mixture. The use of chemical admixtures greatly reduced air content. Shrinkage reducing admixtures reduced the air content within the concrete significantly. Increased percent cement replacement with SCMs increased the workability of the mixtures. When necessary, careful addition ofHRWRA extended mixing times. Increased time in the mixer deflates the mixture and results in a decreased air content. The set retarder increased air content slightly with only an average recommended dosage rate. Unit weight for all eleven mixtures varied due to the fluctuation in air content. When the design "predicted" unit weight was adjusted for the measured "actual" air content, the revised unit weight was reasonably close to the measured value. The L W A mixture did not produce a light weight concrete. It produced a concrete of comparable unit weight to the other eleven mixtures. Some of the mixtures with the highest resulted in the largest unit weight. This is again due to low air content. 197

PAGE 217

Ambient and internal concrete temperatures were average and within appropriate ranges for concrete placement. Temperature is not believed to have played a significant role in this study. Lower w/cm will result in high early compressive strengths and rates of strength and strain development. Increasing the w/cm to 0.44 and Class F fly ash replacement levels up to 30% was beneficial in controlling strength gain. Mixture 5 (0.44/6.5/F A301II) did so, resulting in a comparable rate of strength development to its control mixture but decreased the strain development and 56-day ultimate strength. A low cement content mixture with increased w/cm and fly ash replacement proved to be beneficial. When SCMs are not utilized, a low cement content of 6.0 bags is beneficial. When SCMs are used, increased cement content may be necessary to maintain the same properties. Type G, coarse-ground cement was beneficial to strain and strength at the higher w/cm of 0.42 and low cementitious materials content. At lower w/cm of 0.38 the cement behaved similarly to the control mixture fabricated using Type II cement, developing strain and strength at an average rate.

PAGE 218

A high dosage rate of a shrinkage reducing admixture is extremely beneficial in controlling both the development rate and ultimate strain of the mixture, while maintaining adequate development of ultimate strength at all ages. An average dosage rate of a set retarder only retarded the initial strength development slightly. After 1 day of age, the development of strength and strain was substantially increased. Although the concrete containing the set retarder reached higher compressive strengths more quickly than anticipated, the concrete did not crack in the AASHTO PP34 test and was moderately durable. Table 8.1 shows the 56 days of age compressive strength and penneability results. In addition, the results of the restrained shrinkage test are included The mixture designs batched were used as a basis for analysis and variations and are utilized in developing recommendations to current Class Hand HT specifications. Table 8.2 compares the mixture designs examined in this study with the Class H and HT specification requirements for compressive strength penneability and cracking tendency. 199

PAGE 219

'=' :z::z::z::z::z::z::z::z::z::z::z:

PAGE 220

8 3 201

PAGE 221

A summary of recommended adjustments to the current CDOT Class Hand HT structural concrete follows: Increase maximum allowable from 0.42 to 0.44; Increase maximum allowable cement replacement with Class F fly ash from 20-30%; Incorporate the use of cement replacement with ground-granulated blast furnace slag up to 50%; Incorporate the use of a shrinkage reducing admixture at high dosage rates ; Incorporate the use of a set retarder admixture at average dosage rates ; Decrease cement content to 564 lb / cy when supplementary cementitious materials are not used. 202

PAGE 222

BIBLIOGRAPHY Attiogbe, Emmanuel K.; Weiss, Jason; See, Heather T., (2004). "A Look At The Stress Rate Versus Time Of Cracking Relationship Observed In the Restrained Shrinkage Test." International RILEM Symposium on Concrete Science and Engineering: A Tribute to Arnon Bentur, RILEM, France Atis, CD., (2003). "High-volume fly ash concrete with high strength and low drying shrinkage." Journal of Materials in Civil Engineering. Vol. 15, no. 2, pp. 153-156. Mar.-Apr., Turkey Babaei, K. and Hawkins, N.M., (1987). "Evaluation of Bridge Deck Protective Strategies." NCHRP Report 297, Transportation Research Board, Washington, DC. Babaei, K., and R. Purvis. "Prevention of Cracks in Concrete Bridge Decks: Report on Laboratory Investigations of Concrete Shrinkage." Report No: P A-FHW A-95-004+89-0 1, 1995, Pennsylvania. Bissonnette, Benoit; Pierre, Pascale; Pigeon, Michel, (1999). "Influence of key parameters on drying shrinkage of cementitious materials." Cement and Concrete Research, V.29, pp. 1655-1662, Quebec Brewer, Harold W. and Burrows, Richard W., (1951). "Coarse-Ground Cement Makes More Durable Concrete." Journal of the American Concrete Institute, January, 1951, Title No. 47-25, V. 22, No.5, Michigan. Burrows, Richard W., (2003). "A New Crack-Resistant Cement?" Report Prepared for Presentation to ASTM, June, 2003. Burrows, R.W. "Remarks on the Ring Shrinkage Test." Report Prepared for Presentation to ASTM, Cabrera JG, Atis CD. (1999). "Design and properties of high volume fly ash high performance concrete." Proceedings of ACI International Conference on High Performance Concrete and Performance and Quality Control Structures, Gramado, RS, Brazil; SP-18, 1999, pp. 21-37, Brazil 203

PAGE 223

Cusson, D. and Hoogeveen, T., (2006). "Preventing autogenous shrinkage of high performance concrete structures by internal curing." NRCC-48368, S.P. Shah Symposium on Measuring, Monitoring, and Modeling Concrete Properties, National Research Council Canada, Institute for Research in Construction, Canada. Deshpande, Swapnil; Darwin, David; Browning, JoAnn, (2007). "Evaluating Free Shrinkage of Concrete for Control of Cracking in Bridge Decks." Structural Engineering and Engineering Materials SM Report No. 89, January 2007, The University of Kansas Center for Research, Inc., Lawerence, Kansas. Hadidi, Rambod and Saadeghvaziri, Ala, (2005). "Transverse Cracking of Concrete Bridge Decks: State-of-the-Art." Journal of Bridge Engineering, September/October, American Society of Civil Engineers, Reston, Va .. Krauss P.O. and Rogalla, E.A. (1996). "Transverse Cracking in Newly Constructed Bridge Decks." National Cooperative Highway Research Board, Report 380, Transportation Research Board Executive Committee 1996, NCHRP Project 12-37, Washington, DC. Mindess, Sidney; Young, J. Francis; Darwin, David, (2003). Concrete." Prentice Hall, New Jersey. Mokarem, David W.; Meyerson, Richard M.; Weyers, Richard E., (2003). "Development of Concrete Shrinkage Performance Specifications. The University of Virginia and The Virginia Transportation Research Council, VTRC 04-CR 1, Virginia Perragaux, G.R. and Brewster, D.R., (1992). "In-Service Performance of Epoxy Coated Steel Reinforcement in Bridge Decks-Final Report." New York State Department of Transportation Technical Report 92-3, New York. Philips, M.V. and Ramey, G.E., (1997). "Bridge Deck Construction Using Type K Cement." Journal of Bridge Engineering, American Society of Civil Engineers, Virginia. Purvis, R. (1989). "Prevention of Cracks in Concrete Bridge Decks." Wilbur Smith Associates, Report on Work in Progress. for PennDOT Research Project 89-01 Pennsylvania. 204

PAGE 224

Tritsch, Nathan; Darwin, David; Browning, JoAnn, (2005). "Evaluating Shrinkage and Cracking Behavior of Concrete Using Restrained Ring and Free Shrinkage Tests." The Transportation Pooled Fund Program Project No. TPF -5(051), Structural Engineering and Engineering Materials SM Report No.77, The University of Kansas Center for Research, Inc. Lawerence, Kansas. Whiting, David A.; Detwiler, Rachel J.; Lagergren, Eric S., (2000). "Cracking Tendency and Drying Shrinkage of Silica Fume Concrete for Bridge Deck Applications." American Concrete Institute Materials Journal, V.97, pp. 7177, American concrete Institute, Michigan. Xi, Yunping; Shing, Benson; Abu-Hejleh, Naser; Asiz, Andi; Suwito, A.; Xie, Zhaohui; Ababneh, Ayman (2003). "Assessment of the Cracking Problem In Newly Constructed Bridge Decks in Colorado.", Colorado Department of Transportation Research Branch, March, Report No. CDOT-DTD-R-20033, Denver, CO. Xi, Yunping; Shing, Benson; Xie, Zhaohui, (2001). "Development of Optimal Concrete Mix Designs for Bridge Decks." Report No. CDOT-DTD-R-200111, University of Colorado, Boulder, Colorado, Colorado Department of Transportation, Colorado. Annual Book of ASTM Standards, (2007). American Society of Testing and Materials, Philadelphia, Pennsylvania "Control of Cracking in Concrete." Transportation Research Board of The National Academies, Transportation Research Circular E-CI07, October 2006, Basic Research and Emerging Technologies Related to Concrete Committee, Washington, DC. 205

PAGE 225

APPENDIX A Mixture Designs .. __ 1 -. ;;.t j;;.(SSD)---cf 4$0 2.44 0.0'0 3.15 123 0.$1 o.on Fly .... 2.37 0 0.00 0.000 2.'0 n 0.2) 0.00 2.20 1766 10. 8 4 0.402 2.61 0.80 S4",4 1143 6 0.258 2.63 0 70 24) 3.'0 0 144 0 .065 1.1, 0.0'5 27.00 tOO r .. cteristiC$ .. .. .. .. 14,..4 .. ,.4 ..... .14,,,,,1 O U -0.0004 0 .0$ 0 00125 "A" ___ .". _____ "_ 4$0 Co .. pr ... iy.
PAGE 226

Spr.adsh-;it----------------! ..:. .... 2.3* O.OU 3.15 0.43 0.023 D1 0-00 0.000 2.'0 0.15 0.005 2.20 10.*4 0.402 2.41 UO 1.35 2.43 0.10 3.'0 0.145 1.14 Q.045 27.00 1.00 0.42 14M 5$0 41 4, 4, 0.4' 441 '3 20 1153 1205 251 !HRWRIAE 1.5 HRWRIAE 251 3.0 5U 10.3 0.0 D 1'4.8 133.' 28.5 AEA 28.4 HRWRA 10Q 207

PAGE 227

. 3 c.o"CU!!t ... ... ; ... ... .... ...... ....., ............... .... . .... . ....... ; ...... 3.15 123 0.$1 o .on U7 0 00 0 000 2 .'0 0 23 0 .00' 2 20 1 0 .34 2.'1 0 .30 0 25* B 3 0 .70 3 90 0 144 1.7' 0 .0'5 27.00 1.00 ; 0.33 4$0 123 lUO 1145 1$1 iHRWRIAE 1 0 .HRWRIAU 142 M ..... 3 0 53. 3 .. 0 0 202.2 127.2 20 IHRWRIAE 15.3 208

PAGE 228

[ I P S.G. ....c 3.15 93 F 'BFS 0 0.00 0.000 BFS 20 0.005 17H 0 402 0.$0 7.35 H3 244 0.145 27.00 1 00 S. Co-rno-l'Ititi ...... M4t. z 1 Z 93 Z 0 0 Z 3 5 20 4U Compr
PAGE 229

.. .... : 42$ 2.1. 0 0 $ 1 3 .15 1 8 3 0 .04' __ 'V ____ ___ y_. FI 2.31 : IlFS 0.00 0 000 IlFS 2.90 OM 0 000 UO 10.$4 0 402 2 .'1 0.$ 0 10" 0.241 2 U 0 .10 2'9 4.31 0.1'0 O .O'S 1.1, 0.0'5 21 00 1.00 r=. .. 0 .80 0.10 ... if .. d 42* Compr i cylind ... 11 0.10 fI,,.,h 1$3 RCIP cylinder 0.00 I BFS MOR 0 00 Unit weight 0.25 1154 0.00 10H ... ............. S.lt 0.00 2$0 MOE 0.00 1 0 mMold. 0 00 ......... HRWRIAU 12' S plit Cylinder 0 00 ... ....... ITot.1 2.15 B .. 3.2 Ix 1.15 3 .14 50.1 21.5 !BrS 0 0 0.0 lZ S 32 HRWRII'tE ..I 210

PAGE 230

: .. I-:'-N",.", 3,15 'Fly Alh 1.24 '8FS 0 0 00 0,000 31 O,H O.OOf O,OU Fly Alh i8FS 3 1 17n 1 0 :HRWRIA[ 117 : D 4*.5 FlyAlh 22,4 !8FS 0,0 3,7 214, 5 131.8 34 'HRWRIA[ 14. 4 0,245 4 .31 o.no o,oU 27,00 UO ____ Ib v ______ .. I ___ ___ _ v .. I Fly Alh 2 31 8FS 2,90 2.20 0,*0 U3 0.70 0.00 0,00 0.2 5 0.00 0.00 0 0,00 0.00 0.00 __ v_ ____ ITO!" 2.75 3.30 211

PAGE 231

: ,.7 ",i;Pro ortio. (Ssoy M I P A.C ... 1: )0' 1.55 0,058 ).15 'Fi, .... 0 00 0.000 Fi, .... D7 )0' U, 0.0') 2.'0 0.00 0.000 2 20 10.*4 0.402 Hl 0.80 112. .35 0.254 2 B 0.10 2" 01 0.1'0 0 .0'5 0.065 21 00 1 00 --------. 0 44 .f 1)'. Z 611 z 4. 0 .10 -0.00' .00' 0.10 Fi, .... 0.00 )0' 0.00 Silic" Fu,.... 0,25 1155 0 0.00 111. 0 0.00 2 8 0.00 HFiWRfAE 1 0 0 00 HFiWRIAE '0 0.00 ITot,1 US .1.2 ),)0 Fiy .... 0.0 31. ) 0 0 21 5 -.A __ 1)'. -) 5 0 HRWRIAE 11.0 212 L : 0,2*0

PAGE 232

-. --. 0.$0) 0.70 $., C.", .. rr.,iti.v M .. ,.1 Arh u I.c .. "-,,,rr.t :..; Mirr..D., S.R.A. )0 16.2: 0.5 6.1 M 5.11.".

PAGE 233

, .;;. S .G. '9fS Silk. f' ... "". 1 fl ode ... )18 17" 238 O .OU 3 0 ; 0.071 3.15 1.10 0.041 fl U? 0.00 0 000 9fS 0 00 0.000 UO 100$4 7.58 0.281 H3 HI 0 .141 0.0'5 1.00 '0.0057 tc. I,d' ......... Testi ... 12 __ __ ... RCIP c?lind... 0 tSfS MOR 0 ; R .. k 175' U45 S.lt Ponding 0 MOE 1.0 B mMold. Split Cylind., 0 80 0 70 3.5 ............. ........ 1 :;::.1:.....-+_-:;,:;--/ 49 0 0 0 0 0 1'1.4 3U 14.5 214

PAGE 234

: Mis P.o o.tio. (SSO) ., ..... ..... Mate.ia' P.ope.ties H.hri.1 1 .. lcy) d c .. ..c M .. t,ui.1 S .G. A.C 1 C.m.r.' 2:.*7 1).104 C.II'I.n' 3.15 __ __ 0 .00 0,000 2,'0 .SiliuF' ... "". 0,00 0 .000 iRa-:1c 17H 0,402 0.$ 0 lS.,.4 '0' 0,20 5 1.$1 1*.71 is-an,,JLWA 250 2.21 0 ,03 2 0.7 : W41t., 237 3.10 0,141 Air 0.045 1.74 o .on 27 .00 1 .0 0 1 u/c --, ---_--..._y __ : Moist.n Co.teat ( .aditio ... Rocll ... lGlh4,. .. ",_19R.l. IGn4 .... ,.ut [rackl." raclc+ ... nut ,:;. 21 ... 1.r4l\4ft'1oc %)1 .nckn'lc 0 .12 0.10 -J,yut.r.,.4 0 ,0002 0.007 Batc' Yeig'ts (,d") 544 SFS 0,70 0,00 0 00 Silie.Fllm. Rack 0,00 __ ____ ______ __ """''' S.II Po:;:.n",d",in"'-r-__ +_.:.......;t--....:;=_+ 1754 0,00 '07 S-an4lWA 24' 0,00 0.00 MOE t Wo,., 250 Split C lindo, 0 00 1 0 fl".'cl.ollt 147 1 Batc' Yeig'ts (ft" 0,'5 3.3 5 !B-al::ch.li: .. ),4
PAGE 235

.. ,Mi. Pro ortio. (SSD) H.tl>ri,,1 W.i,,"t III-ley' Y.I.,,,,. d c,....".,.!t H4 2.U 0 0 .00 0 0.00 SiliuF .... m. 0 0.0 0 ,RD-ck 11" 10.*4 1270 7.74 H"'''' 0 0.00 ,W.to r 231 3.3 0 o.on -------27 00 j._--_. 1 ul. 0.42 142.1 lJah"fl'e6 0.10' 0 000 0 000 0 000 0. 402 0.2$1 0.000 0.141 0.0'5 1 .0 0 S.G. A.C Pvt:,a. C .. "" .. 2 31 .. __ -L __ __ s ..... ,.1. 1'1.111:. ( % F'Iy Alh ;.1,) () () Moist .. re Co.t.,.t (Tr .. ditio.",1 Roc' ... t.n. 1 .... ,. ..... racle .or. toll.7' r.ck+ nut .2:ilt.I' 1.15 0 0045 !,.-c:k",< Z 0.45 -o.ons . Testi. B",tc. Yei ... ts 1 Compr<:::::::::iyt ( l indtr:::: 0.70 iCo>"'.ht 5U RCIP CV 0.00 MOR 0.00 Unit ht 1 0.25 0.00 FI.ck S.lt Pondin 0 0.00 1275 MOE 0. 00 0 Bo.", Mold. 0.00 .!W.t., 237 S plit Cj lindo, 0.00 'HRWRI ... [ 1 .0 lc ... \ iHRWRI ... [ 1U ... ........ B .. tc. Yeig.ts 3.3 5 !. 'Batd",iu 3 4 cf !C.",.r.t 70 ,0 0 0 0 0 0,0 !R.ck 213.3 S.",4 15*.2 0.0 W.tll>r 2', 4 20 7 "'l iHR.WI! ... 251 .",l

PAGE 236

--..j LABORATORY REPOlq .. (:",(;'-';1 8 SAMPLO BV: PROJECT ; w.;dlaneo", \/\,(,61,,1>1 P R OJEC1 NO : 1,1;",,"" A.'STM C F __ --_. _-_. ._._ -". .......... _-_._--------iSAliPU; ; :',,"s,H!.;t1 .;.:.o"T.'f,Y4 t,; Te)b (.\.:jn.i (; I() M Sul.3tc; AfHMC 117 I.e 136, Mf>H H1 T "r ..113m (; iU.Jurt.Sft'TO B",. 5 C ":",,, ,: a."' ... ( .5M'J. ----:-----r-;;h.:J uHA1JltH,:a 'fVE ltiH7 'RE ;lID:. :, 'RIG I.;" n,; ; ( {.F l'ER 1 ____ AS1 .. _"0 _______ ___ _ ____ __ ,..=,.",,., 0>,.,. __ .__ .. 1 1 W WI) rO) ::. 1 .. 1i) 8t; 1\X1 I r J O PUr\odtJ8 6 loll! t l { .4 --;;-$ Itt If. ... G(' ocU':<. --1 1 e I 123. t ___ 11 __ ,,3f.) 2'; 1 :25 -tiD "weu1 :;; 00 1.41 .... r"'lG __ ,_ 00_ J_" l). <:.1), AASHiO l,t. :01L,i,t> i i '-.--. "1.1, 8r;,."..-:_4 ..': &Ui\;'t; S ill:. f-I ('l) 8 0 ;> tTl 0 OJ

PAGE 237

00 ',H OC!,;CSlIP liON C U E N f r SOURCI i6 r '\tllL:f1 SAMPLE[) 8Y: Clo:lrl Wo.THt I'I
PAGE 238

f,,}t ..:wrt \lij}j"k:,,:! Pn ASfM C 1 :6[1 C P-L C :(II P<.!h:n.t!i :ll .::>l'M .(" J 3 J3 F,n e \ \:,d : t..;t ",, :. 1:S Fi'e,f\I!t:. i I :' the tlklth !;" t :th lltJ) In-n n lnt:' t>n 10-08 The tt.ii!Cll Lt + Jfi.IOO ; \::iT \ i '.l[f;11'iiJ'x; ,jcid(r 'IJ'.Il' :,:'(p!f.n:"i(,lr\ .' n .w", '.' l i :: i;'0.1& i:i' ,l"'t\! UOO5 L' 219

PAGE 239

:rrH4nq. Rr...C.CTl'Ji.'I"-t 0 f ';-':''''''''H .. t;;;.:a:o i:!,.:,t:.H.iENT:5' GI!ffififil Gi1VJ M':tJ;fT; .--: ?7(); 4 ZJ:';t:(i:S 1 s ..... : If,VJI ..... ... IH HO (gj .. f+v IlN,<'\. 'loU : 0 "':;;': 220 -..J

PAGE 240

lABORATORY TEST REPOR T ,.:..;n:N n..t.t Q :t. A!,;H4 'fi'1 :'1;.' :'_ GGAE .c;.:.\TE_:l "'-U<;Jd.JS 1.:'r1 I; 1 U H DoYt /"';;;';;.. A -<1);.111-4 l;l-:{ H <1 OS*' 1 -6 ..(;';OU -Dil'U; { tLfH 4 jj';o.!:l .. i}.1H 0 -<.Hi1;: i.L(":2 0-.!42 221

PAGE 241

222

PAGE 242

lertaJ Sample Numbe/, Simple D.",'' TEsrs C("IIQC S(l1061l0 II Moy 2007 TESTS 'JIo O.'de '10 Iran (FI203). % )0.1. P6 R.pott DatI : Simple Source : Slim of 5102. A1203. FeW3. % Calcium Oxide (CaDI, % % Sullu, Trto.ide (S03), % SodtlJrn O.lde (NIWI. % 'Ao 28 7M 1 .74 OJ) 2 .29 70.1Ill0.0 min O/SD. Omi, TOIaI AfIIllIU 'A. (ItS 0 mIx. Ilm ... ContinI. % 0 .11) lio LoIS on Ignition % 11. 16 6 .0,na.. 0 m ... Amounl Rotllntd on No. 325 Si,,,., % lS J4,,"I.' Gravity 2 79 Autoel ... Sound"",". '4 0 0) "'.. '18,nl'. SAl, Ponland Cement dIVI, % of Control 9H H 'nln.' 7 .. 5.<\1, WIllI Ponllllld Cemental 28 d.ve. 'Ar of Control 9&.8 7l,nin. t ,nlo .. Wate, Req,,,ed. % Control 93. 105 n'n. IfS",.,.. M .... "'1M C III Ind AA8HTO M 215. C).I. C Cia .. Fly from mtellihe 1Wq,'remeolll of J lftll comptilnl:'e I \i '/-'<,<" .. SA. A//TONIO. TEX..., Ap"rovid !Iv:

PAGE 243

Borel Material Tedmologtes Sample Oaln TESTS ClilllQC $-07062501 I 2001 O_kIt 0/. (c.O). MIgn.alum 0.''''' % Pot.uium % Tolel AIIIIII&I (II % / Moitlure Content. % ,/ loas % / / CeMenI ,I daya, 0' eMtlol a days, 01 % Conll r RESUlTS 24., 4 .24 83.7$ 9 .71 0"7 0 .99 0 .09 0.6' 2 .S1 110. 89.3 RepettOII.: 70.0150. 0 min. S.Om"". 3 .0'n .. 6.0m.,. 34 min. 75 SIlO 007 7 .0m.Omi tmax. ... 14.0 MI ... ASTM C ttl AA TO 2t1, PDOT 'D. MOOT 'PICIllcIIlOn' lor C .... Fly "'" .. Approved By, -:)Pfd!iIII$1 Mlfori,'-Miaif' : 'SNI 100 SAN ANTONIO. TIXAa 21U.O 224

PAGE 244

J 30 19 MftY 2007 elliS QC S l lcon % 52.30 'Yo 25.1. % of % 82. 2 1 min. ICIO), % 9 n Oxide (MgO), 2 0t 0 .44 Sodium Oldde (NIZO) 0/0 0.71 % Tolar (II 1 .34 'Yo 0 .44 MollltA Conten\ 'Yo 0 02 3 0 Loll on IgnHlOn % 0 ,70 R,ta lnacl No, 326 24,64 mM. S."cIRc; 0 .01 0 8 Portland at % Control 8H wllh Clmltit I' % Control 15 Control 9B lOS Milia ASTM C .11 alld MaHTO M 2." I'DOT S.otlOn nt, SCDHPT MOOT 'Plclfleltlont Clan Iho 'd.y or 21 dtr S-oll\ Aenvlly Index
PAGE 245

F. n.-a.t' 19 ...... ... IaIi""af. "",ddi
PAGE 246

t::: ()t'll tVlt !'Xatch ll>ac .... 19 bk ... it1o ()Dc .. 1 ..... 6ey_odd
PAGE 247

':\, ;" ::c Doa. ...... AT6(J _c. ..... 11'0 finio .... d .. rod.n_. of c....,,... .... ""''''I1!I.ep''' __ '" =uue1y C c __ ""0lIl..-1. all 111 .my of i< ........ e'l'-or .. dljoclod '" tlor IPpO,",*o;a 0IlI. .-.-.. AT60 .. brao I'ani",. c_ f"'lOr .... law siamp c--... "",ire ''''Yejf1tdly ""'r ..... .. piut:iUu'ly ... 10 Jem_ ....... ......uiII rMI_e ",.tao "ffi .. &lid ......... AT60 .odiiilio ....... ;Jpem..t lIo\"!af lir Add .... ntao...,a1.., 1""",,01 i-.:la.:Ii"!!IPa:lficmit. p-cn. ..-;.J <.".,.., "'_ .... be "" ..... df""_afabrdlemic.h4mixi_ .. lioe*>n .a ...... con,,,,," 1_ ""'''odt''.reqairod daoofe"I t c .. _e tile
PAGE 248

C om oat Other ATro .. ill, ""'. Gnce >dmillllRO .. Ihoy .... Jdd ... .",... ""<01)' ..... ro:"",mood'" .... n..., ... ATro be...u.d '0 caac ..... 041:1, ........ <2 !>y ...t. ......... "' .. "'61 bolllOdifl",._ ... 4>0 .... !'\No ..... ()nee Tloooiaoi TB-llIIC, i_e,,_ 0"" ....... &10....., far r ......... ""' ....... .-. Don ....... ATro 4>oulitllalbelddoddw-.tdy'obNred ... .... enn ...... Wbe ;>:ri>1med chll,. .. old",.., ....... ""' .... to ;>:ri> .. for piduc:.,. D ... by""'creo:t..a 'nd: ...tlullpl (lOSL ) Don, ... AT8!l .""-Id be prtICO:"" \lat.f a:_. *""",,",,, mrca..,. t.Te hdI MSDS s h_dlio, pr
PAGE 249

Il-RASF The 0.0. ..... ......... .. --... ....... -......... ..........,._..,... -...-_ .... --Ior ...... -.. ---.. ........ ,n ..... -----..... CIIIYIII ..... --.. ..... ..,... ,....A.a_ -_ .... -111: ---.-,.,..... .. ..... ."._ .... a .. .... ,.".. -_ .... --..,... ............ ........ ...... '.Ia't ...., ......... ___ ...... JJ'_...-.----_ ... ...... I 1 -q 1 1 .... ... _Ctw........ ......... .............. ................. -... ...................... .......... ___ .......,-...... ..... .................. ... ...... ...., ............. ............ ....... .... -..-01_.11 .. ........... .. __ ........... _--' De ....... ..........,-.... --.. ................................. .. llp -_ ..... n.....--. ..... .. .... __ ........ _."""""....._.----....-go--

PAGE 250

Data: bialdI ,'II ,. ( -"'
PAGE 251

fr.e fat t ___ ..... __ I00m ...,.-... !Qr BetV!fiI& 1M-"'_1wbtIJ 11_-",,- __ ...w", ... ct __ Per1ornwICI!! _lIanno. __ ... "9""" __ _"""''''''' __ .... 1/:1 ... 8_ ..... _10'''.'''.'''''''' ..-_1o. 1!>4_ .... .._,.. __ __ 11.pIIS .......,..__...-..... AtsnII C>4lWC'_1'fpM ""'_l00XR""'''_ ,\1 __ ... 'OO:>:R __ ._ ... -.c ___ ",",""IJIholmo c.........., 00_ ..,-_ ....... pooIod.I ____ 232

PAGE 252

Product Data: P02:2:01J rH'"1 OJ) x n ,--,._ ... __ .... -umg "' __ __ "';.o .... M6Fc..._m-. ___ __ N10XR ___ __ ;(11_"'. ___ bc _____ "'t. ___ ... __ __ __ StotIIge and ti8ondIlt'9 3t"" __ ___ "1bi c:t.mt ..,JIIool ...... _IOOm4.:l"' .. "' .. _"-_ ...... -rOOm_"I\j ... <:2tl!IlJ "'_ .. __ .. __ Irid'ormIIdoIII -.... """-""" """,, __ Irt "'1>1..--'-eu_s---' ____ .. ...... .. t.,...... __ ........ ... Master Builders c...,_____ ....... : .. 233

PAGE 253

APPENDIXC DOT Survey 1. Crack RHinant Conuet. for Use-In tkIdge Decks ... """ "91 aadIIn4 ..... .... ... ... ,.. ()O;" In, liI94 2. Contad Information ...... 3. Bridge D.de Concrete: Cradunq O<:ctUTenc. and TesUIU; 2. .....

PAGE 254

,. ...... $-" ... ..... ... -t; ...... t_. >:_ tJo __ ... .t dOet tM Mint f'Ot deeU 1ft It.tate tl'l*
PAGE 255

WNt any, NW MMe to JUte Department of dedi mlwtu"e .s.t19n_ haY. rftwltM t. S tillS stat. Department Tratt."wtaUott brid9* dec" ( on<:rle th .. 1iItlllud ... fink ...... admtxtu,.,. (S.It.A.)1 11 __ .... ............ __ ... .... .. _, .. -...."..... __ ..... Y'OUr state De.,.,tment th.l.t COf'llpI!ftNtiftg (erneot (sec)? B::-10. Willett tIM matetiall bdow. WHJI IIK'udM 'ft 'M 1),14,. dedi COIKtet4 me.tuN sect crKklnl 01 yOW .... tekp.tftmeftt rr.n."wtaUon' .". de<:k.1 c ..... \: .... .... e-..'", --he:,........ o o ( .. o o 236 'Ow ...... C C

PAGE 256

11. Wl'lk:_ IMterlatlkJow wile. iIKWect III tM brl4" dec_
PAGE 257

Depar1m tnt ',:on1l ct ClIlpar1m tntContic .al ICl.J1l Dr.JDrJooutJILlk n Jl(Izona Pl:lfllaz'SI1.c<.:III .Jl/1!.anlu IIl!\ltall3l 1I1WlM JXtII OIlIfCrnll 11'01 art ....... COlorailo Ira. rd>t."",. Olnn.cllcut 'IIellll3l. .ct" c l rlUl'. flol r<:be rtrablalllDU. ct.1' Glawera (,eorgla ,toll 'm .. It't, .. ", .." pal lIIUJrLCi1UtlltJJil .I' klaho Milt rntlrra.-m" l1.1' Ilr .... ". Vania' h"''''t. md. 1l Ma"athuut1l ma.I' Mlchlaan nlllud4!m 1:1 klal .QJII 1.1Iu(ld)ml::ll:JiI. IPI Mlnl1l 10111 mus i.1a1.doIl314!oot.1nm I.U Mluh.IPQ) -"".11 Mlnoun laP" c lcHlfr l1n. 1 e 1l I.vada Him IlIhlra C'.as=Hf.ldotl1n .II.U I.'. MIlito IHTOII)I1n .Im n Slenna'.Pe.r.olllllmUtlJ1n.lfn.U UIW'fOnt ,alilllp.aI"IW.ooutilil .1V. n Itlrth 01 ru 1111 : rIlOI Itlrth Clkota 010 oIIJaltlallllooU1nDI. 1I TInI. 11 utah I ,1:>1 "irmont MU .He(tJ!, 4!;tiJt lit" l;lrmnla t l i lralnla a1tIUIld:lUtII! _.11 'fI,con,ln

PAGE 258

APPENDIX Photographs of Cracked Restrained Ring Shrinkage Test Specimens Mixture #1 (O.38-6.8-F A20-SF5-II) did not exhibit surface cracking Mixture #7 (0.44 / 6.5 / BFS50/II) cracked at 32 days of age, and an average of approximately 90micro strain (No Surface Cracking) Mixture #8 (0.44-6.0-F A30-SRA-II) had not cracked at 56 days of age, and an average of approximately 73micro strain (No Surface cracking) Mixture #3 (O.38-6.8-FA20-SF5-G) Ringl 239

PAGE 259

#2 (0.42/6.2/F A 16/SF3 .5111),

PAGE 260

#4

PAGE 261

Mixture #5 (0.44 / 6.5 / FA30/Il) Ring #1 242

PAGE 262

#6 (0.44 / 6.5 / F A30/ SF51II),

PAGE 263

Mixture #9 (0.44-6.0-FA30-RET-II), Ring #1 244