Application of the Maturity Method for the Prediction of Early Strength of Concrete under Various Curing Conditions

MISSING IMAGE

Material Information

Title:
Application of the Maturity Method for the Prediction of Early Strength of Concrete under Various Curing Conditions
Physical Description:
1 online resource (159 p.)
Language:
english
Creator:
Kwon, Oh Hoon
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Civil Engineering, Civil and Coastal Engineering
Committee Chair:
TIA,MANG
Committee Co-Chair:
GLAGOLA,CHARLES ROBERT
Committee Members:
NAJAFI,FAZIL T
MUSZYNSKI,LARRY C

Subjects

Subjects / Keywords:
concrete -- maturity -- prediction
Civil and Coastal Engineering -- Dissertations, Academic -- UF
Genre:
Civil Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Replacing concrete slabs on the highway facilities requires a contractor to close the lane until the replaced concrete slab has enough strength as a pavement. In order to test the strength of the replaced concrete slab to determine the time to open to traffic, 4”×8” cylindrical specimens cured under identical curing condition are made and tested. However, the strength of the concrete specimens does not accurately represent the actual strength of the in-place concrete. Furthermore, the compressive strength differences between different locations of the replaced concrete slab were founded to reach as high as 30% at the same curing time. For these reasons, maturity method was proposed to determine the strength of replaced concrete slab as an alternative of testing 4”×8” field cured cylinders. In order to determine the most appropriate testing procedures to be used to obtain accurate strength prediction of the concrete used to slab replacement project, various possible factors which could affect the maturity strength prediction including maturity systems, specimen size, fresh concrete properties, curing conditions, maturity functions, and modeling functions were evaluated throughout the laboratory experiment. Investigation was also made to evaluate the effectiveness and reliability of the maturity strength prediction throughout multiple field studies. The laboratory generated maturity-strength curve was compared to the one generated at the project site to validate the strength prediction of the maturity method. Strength predictions for the different locations of the replaced concrete slab were made to find the weakest location at the same curing time and thus the most appropriate location for installing temperature data logger was determined. The results from this study shows that the maturity strength prediction using the proposed testing procedure is a very reliable and convenient method for predicting early-age strength of concrete used in slab replacement project.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Oh Hoon Kwon.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: TIA,MANG.
Local:
Co-adviser: GLAGOLA,CHARLES ROBERT.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0046301:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 APPLICATION OF THE MATURITY METHOD FOR THE PREDICTION OF EARLY STRENGTH OF CONCRETE UNDER VARIOUS CURING CONDITIONS By OHHOON KWON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FUL FILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 Ohhoon Kwon

PAGE 3

3 To my all family members

PAGE 4

4 ACKNOWLEDGMENTS First and foremost, I would like to express m y deepest appreciation to my advisor and committee chair, Dr. Mang Tia, for his constant encouragement, valuable guidance and sincere support at University of Florida (UF). Without his guidance and support, this dissertation would not have been possible. M y sincere gratitude is also extended to my committee co chair, Dr. Charles Robert Glagola, for his guidance and help. I would also like to express my sincere thanks to Dr. Larry C Muszynski for his constant guidance as a co principle investigator of the s l ab replacement project and Dr. Fazil T Najafi, for his generous support as an IRF member as well. In addition, I would like to thank my colleagues and the staff of the Department of Civil and Coastal Engineering at UF for their kind support during the pas t four years. I also thank my friends in the Korean Student Association at UF for their unceasing encouragement and sincere friendship. Special thanks to the Florida Department of Transportation (FDOT) for sponsoring the research and providing both the fin ancial and technical support that made this dissertation possible. Last but not least, I would like to express my sincere appreciation to my parents, to my family members, especially to my wife (Hyunjung Kim), my young daughter (Soohyun Kwon), and a baby on the way due in January, for their understanding, patience, encouragement, support and endless love.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 11 ABSTRACT ................................ ................................ ................................ ................... 17 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 19 1.1 Background ................................ ................................ ................................ ....... 19 1.2 Problem Statement ................................ ................................ ........................... 20 1.3 Hypothesis ................................ ................................ ................................ ........ 21 1.4 Objectives ................................ ................................ ................................ ......... 21 1.5 Scope of Study ................................ ................................ ................................ 21 1.6 Research Approach and Methodology ................................ .............................. 22 1.6.1 Literature R eview ................................ ................................ ..................... 22 1.6.2 The First Set of Experiment s ................................ ................................ ... 22 1.6.3 The Second Set of Experiment s ................................ .............................. 23 1.6.4 Field Study ................................ ................................ .............................. 23 2 LITERATURE REVIEW ................................ ................................ .......................... 25 2.1 Slab Replacement Project ................................ ................................ ................ 25 2.2 Strength Prediction with Maturity Method ................................ ......................... 26 2.2.1 Nurse Saul Maturity Function ................................ ................................ .. 27 2.2.2 Arrhenius Maturity Function ................................ ................................ ..... 27 2.3 Comparisons of Maturity Functions ................................ ................................ ... 28 2.4 Limitations of Maturity Method ................................ ................................ .......... 30 2.4.1 Effect of Curing Temperature ................................ ................................ .. 30 2.4.2 Effect of Moist Curing ................................ ................................ .............. 32 2.4.3 Effect of Fresh Concr ete Properties ................................ ........................ 33 2.4.3.1 Water to cement ratio ................................ ................................ .... 33 2.4.3.2 Air content ................................ ................................ ...................... 34 2.5 Functions for Modeling Maturity Strength Relationships ................................ ... 37 2.5.1 Modified Exponential Function ................................ ................................ 37 2.5.2 Modified Hyperbo lic Function ................................ ................................ .. 38 2.5.3 Logarithmic Function ................................ ................................ ............... 38 3 PREPARATION OF CONCRETE SPECIMENS ................................ ..................... 40 3.1 Introduction ................................ ................................ ................................ ....... 40

PAGE 6

6 3.2 Concrete Mix Designs ................................ ................................ ....................... 40 3.3 Preparation of Concrete Ingredients ................................ ................................ 41 3.3.1 Cement ................................ ................................ ................................ .... 41 3.3.2 Aggregate ................................ ................................ ................................ 42 3.3.3 Admixtures ................................ ................................ .............................. 43 3.4 Temperature Sensor /Logger Installation ................................ ........................... 44 3.5 Tests for Fresh and Hardened Concrete ................................ ........................... 45 3.5.1 Tim e of Set ................................ ................................ .............................. 45 3.5.2 Slump ................................ ................................ ................................ ...... 46 3.5.3 Air Content ................................ ................................ .............................. 46 3.5.4 Unit Weight ................................ ................................ .............................. 46 3.5.5 Temperature of Concrete Mixture ................................ ............................ 47 3.5.6 Compressive Strength ................................ ................................ ............. 47 4 THE FIRST SET OF LABORATORY EXPERIMENT S ................................ ........... 49 4.1 Laboratory Experiment Design ................................ ................................ ......... 49 4.1.1 Main Objective s ................................ ................................ ....................... 49 4.1.2 Maturity Systems ................................ ................................ ..................... 49 4.1.2.1 Humboldt c oncrete m aturity m eter ................................ ................. 49 4.1.2.2 Command Center m aturity s ystem ................................ ................. 50 4.1.2.3 Intelli Rock m aturity s ystem ................................ ........................... 51 4.1.2.4 COMA m eter ................................ ................................ .................. 52 4.1.3 Curing Conditions ................................ ................................ .................... 54 4.1.4 Concrete Specimens ................................ ................................ ............... 54 4.2 Evaluation of the Effectiveness of Maturity Syste ms ................................ ......... 55 4.2.1 Comparison of Recorded Temperature Histories ................................ .... 55 4.2.2 Comparison of Accuracy of Temperature Reading ................................ .. 60 4.2.3 Comparison of Equivalent Ages Generated by COMA Meter and Intelli Rock Maturity Meter ................................ ................................ ............. 63 4.2.4 Effects of Embedded Temperature Sensors / L ogg ers on the Strength of Concrete Specimens ................................ ................................ ................. 65 4.2.5 Evaluation of Different Maturity Systems ................................ ................. 67 4.3 Evaluation of the Effectiveness of Different Specimen Size s ............................ 68 .. 68 4.3.2 Comparison of Str .. 70 4.3.3 Effect of Specimen Size on the Maturity Strength Relationship .............. 71 5 THE SE COND SET OF LABORATORY EXPERIMENT S ................................ ...... 73 5.1 Laboratory Experiment Design ................................ ................................ ......... 73 5.1.1 Main Objectives ................................ ................................ ....................... 73 5.1.2 Maturity System s ................................ ................................ ..................... 73 5.1.3 Curing Conditions ................................ ................................ .................... 73 5.1.4 Concrete Specimens ................................ ................................ ............... 74 5.2 Fresh Concrete Properties of the Second Set of Experiment s .......................... 75 5.3 Evaluation of Maturity Functions with Test Result s ................................ ........... 78

PAGE 7

7 5.3.1 Introduction ................................ ................................ .............................. 78 5.3.2 Parametric Study on Nurse Saul Maturity Function ................................ 78 5.3.3 Para metric Study on Arrhenius Maturity Function ................................ ... 86 5.4 Evaluation of Functions for Modeling Maturity Strength Relationship ............... 9 3 5.5 Eval uation of Curing Environments ................................ ................................ ... 96 5.5.1 Comparison of Maturity Strength Plots of Mix #1 ................................ .... 96 5.5.2 Comparison of Maturity Strength Pl ots of Mix #2 ................................ .... 98 5.5.3 Comparison of Maturity Strength Plots of Mix #3 ................................ .. 101 5.6 Evaluation of Effect of Fresh Concrete Properties ................................ .......... 103 5.6.1 Compressive Strength Prediction ................................ .......................... 103 5.6.2 Effect of Variation of Slump on Maturity Strength Plots ......................... 106 5.6.3 Effect of Variation of Air Content on Maturity Strength Plots ................. 109 5.6.4 Effect of Variation of Unit Weight on Maturity Strength Plots ................ 111 5.7 Suggested Testing Protocols for Generating Maturity Strength Curves ......... 113 5. 7 .1 Preparation of Concrete in the Laboratory ................................ ............. 114 5. 7 .2 Testing of Fresh Concrete ................................ ................................ ..... 114 5. 7 .3 Instrumenting Concrete Specimens ................................ ....................... 115 5. 7 .4 Curing of Concrete Specimens ................................ .............................. 115 5. 7 .5 Testing of Concrete Specimens ................................ ............................ 115 5. 7 .6 Development of Maturity Strength Relatio nship ................................ .... 115 6 FIELD STUDY ................................ ................................ ................................ ...... 117 6.1 Testing Plan for Field Study ................................ ................................ ............ 117 6. 1.1 Overview ................................ ................................ ............................... 117 6.1.2 Temperature Measurement s from the Instrumented Slab s ................... 118 6.2 The First Field Study ................................ ................................ ....................... 120 6.2.1 Overview ................................ ................................ ............................... 120 6.2.2 Development of Maturity Strength Curves ................................ ............. 122 6.2.2.1 Laboratory g enerated m aturity s trength c urve ............................. 122 6.2.2.2 Field g enerated m aturity s trength c urve ................................ ....... 123 6.2.2.3 Maturity s trength p lots of p rot ection s pecimens ........................... 124 6.2.3 Validation of Maturity Strength Prediction ................................ ............. 124 6.2.4 Comparisons of the Strength Prediction s at Differe nt Locations of Concrete Slab ................................ ................................ ............................. 129 6.3 The Second Field Study ................................ ................................ .................. 131 6.3.1 Overview ................................ ................................ ............................... 131 6.3.2 Development of Maturity Strength Curve ................................ .............. 132 6.3.3 Validation of Maturity Strength Prediction ................................ ............. 133 6.3.4 Compa risons of the Strength Prediction s at Different Locations of Concrete Slab ................................ ................................ ............................. 137 6.4 Curing Time Adjustment for Field Generated Maturity Strength Curve ........... 140 6.4.1 Overview ................................ ................................ ............................... 140 6.4. 2 Proposed Curing Time Adjustment ................................ ........................ 140 7 CONCLUSIONS AND RECOMMENDATIONS ................................ ..................... 145

PAGE 8

8 7.1 Summary of Findings ................................ ................................ ...................... 145 7.1.1 The First Set of Laboratory Experiment s ................................ ............... 145 7.1.2 The Second Set of Laboratory Experiments ................................ .......... 146 7.1.3 Field Study ................................ ................................ ............................ 148 7.2 Conclusions ................................ ................................ ................................ .... 149 7.3 Recommendations ................................ ................................ .......................... 149 APPENDIX TEST RESULTS ON THE MIX #2 IN THE FIRST SET OF EXPERIMENT S .............. 151 LIST OF REFERENCES ................................ ................................ ............................. 156 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 159

PAGE 9

9 LIST OF TABLES Table page 1 1 Parameters co nsidered in the study ................................ ................................ ... 22 2 1 Effect of concrete materials and production practices on air entrainment .......... 35 3 1 Mix #1 used in both f irst and second set of experiment s ................................ .... 40 3 2 Mix #2 used in both first and second set of experiment s ................................ .... 41 3 3 Mix #3 used in the second set of experiment s ................................ .................... 41 3 4 Physical properties of the Type I/II cement. ................................ ........................ 42 3 5 Physical properties of the fine aggregate ................................ ........................... 42 3 6 Physical properties of the coarse aggregate ................................ ...................... 43 3 7 Admixtures used in the laboratory experiments ................................ .................. 44 3 8 Curing times for testing compressive strength ................................ .................... 48 4 1 Summary of the observed characteristics of the different maturity meters evaluated ................................ ................................ ................................ ............ 67 4 2 ............. 70 5 1 Fresh concrete properties of eight batches of concrete Mix #1 .......................... 75 5 2 Fresh concrete properties of eight batches of concrete Mix #2 .......................... 76 5 3 Fresh concrete properties of eight batches of concrete Mix #3 .......................... 76 5 4 Selected concrete batches for the parametric study ................................ ........... 79 5 5 Results of regression analys e s relating compressive strengths to equivalent age of Mix #1 using three different modeling functions ................................ ....... 94 5 6 Hyperbolic trend lines and predicted strengths calculated by MATLAB program for Mix #1 ................................ ................................ ........................... 104 5 7 Hyperbolic trend lines and predicted strengths calculated by MATLAB program for Mix #2 ................................ ................................ ........................... 105 5 8 Hyperbolic trend lines and predicted strengths calculated by MATLAB program for Mix #3 ................................ ................................ ........................... 106

PAGE 10

10 5 9 Recommended time frame for the laboratory concrete ................................ ..... 114 6 1 Mix design used to both first an d second field studies ................................ ...... 118 6 2 Compressive strength test results and corresponding equivalent ages for the three batches of concrete ................................ ................................ ................. 126 6 3 Results of regression analys e s for two maturity strength relationships with modified exponential function ................................ ................................ .......... 127 6 4 Recorded time for preparation of batches of concrete ................................ ...... 129 6 5 Compressive strength test results and co rresponding equivalent ages for two batches of concrete ................................ ................................ .......................... 134 6 6 Results of regression analysis f or two maturity strength relationship with modified exponential function ................................ ................................ .......... 135 6 7 Recorded time for preparation of batches of concrete ................................ ...... 136 6 8 Fresh concrete properties of the three batches of concrete ............................ 141 6 9 Maturity strength predictions made by different maturity strength curves ......... 144

PAGE 11

11 LIST OF FIGURES Figure page 1 1 Schematic diagram for research approach ................................ ......................... 24 2 1 Process of slab replacement project ................................ ................................ ... 25 2 2 Diagram showing concept of the maturity method ................................ .............. 26 2 3 Comparisons of age conversion factors for both maturity functions ................... 29 2 4 Comparisons of maturity strength curves developed by Nurse Saul maturity function under different curing temperatures ................................ ...................... 30 2 5 Comp arisons of maturity strength curves developed by Arrhenius maturity function under different curing temperatures ................................ ...................... 32 2 6 Effect of water cement ratio on the compressive strength ................................ .. 34 2 7 Relationship between air content and 28 day compressive strength .................. 36 3 1 Pre installed maturity sensors /loggers ..................... 44 3 2 Humboldt A cme penetrometer and mortar container ................................ .......... 45 4 1 Humboldt System 4101 concrete maturity meter with thermocouple wires ........ 50 4 2 Command Center maturity temperature loggers ................................ ................. 51 4 3 Temperature logger and hand held reader of the Intelli Rock maturity sy stem 52 4 4 COMA meter ................................ ................................ ................................ ....... 53 4 5 Temperature laboratory condition ................................ ................................ ............................ 56 4 6 Variations of temperature measurements from same temperature sensors /loggers cured under ambient laboratory condition ................................ 56 4 7 Tempe rature F environment control chamber ................................ ................................ ............. 57 4 8 Variations of temperature measurements from same temperature sensors /loggers cured in 113 F envir onment control chamber ........................... 58 4 9 Temperature curing tank ................................ ................................ ................................ .......... 59

PAGE 12

12 4 10 Variations of temperature measurements from same temperature sensors /loggers cured in standard curing tank ................................ ................... 59 4 11 Different temperature sensors /loggers placed in ice water ................................ 61 4 12 Temperature reading on a mercury thermometer in ice water ............................ 61 4 13 Temperature time plots as recorded by different maturity sensors /loggers in ice water ( replicate 1) ................................ ................................ ......................... 62 4 14 Temperature time plots as recorded by different maturity sensors /loggers in ice water (replicate 2) ................................ ................................ ......................... 6 2 4 15 E quivalent age reading for COMA meter ................................ ............................ 63 4 16 Equivalent ages from COMA meter and Intelli Rock maturity meter under three different curing conditions ................................ ................................ .......... 64 4 17 maturity sensors /loggers ................................ ................................ .................... 65 4 18 Rock temperature logger ................................ ................................ ................................ ................. 66 4 19 ambient temperature as recorded by Intelli Rock temperature logger ................ 68 4 20 ambient temperature as recorded by Command Center temperature logger ..... 69 4 2 1 ambient temperature as recorded by Humboldt maturity system ........................ 69 4 22 Comparisons of compressive strength time plots for specimens cured under ambient laboratory condition ................................ ........ 71 4 23 Maturity ambient laboratory condition ................................ ................................ ............................ 72 5 1 Air content versus mixing room temperature for Mix #2 ................................ ..... 77 5 2 Nurse Saul maturity strength relationships with datum temperature of 5 C for Mix # 1 ................................ ................................ ................................ ................. 80 5 3 Nurse Saul maturity strength relationships with datum temperature of 0 C for Mix #1 ................................ ................................ ................................ ................ 81

PAGE 13

13 5 4 Nurse Saul maturity strength relationships with datum temperature of 10 C for Mix #1 ................................ ................................ ................................ ........... 81 5 5 Nurse Saul maturity strength relationships with datum temperature of 5 C for Mix #2 ................................ ................................ ................................ ................ 82 5 6 Nurse Saul maturity strength relationships with datum temperature of 0 C for Mix #2 ................................ ................................ ................................ ................. 83 5 7 Nurse Saul maturity strength relationships with datum temperature of 10 C for Mix #2 ................................ ................................ ................................ ........... 83 5 8 Nurse Saul maturity strength relationships with datum temperature of 5 C for Mix #3 ................................ ................................ ................................ ................ 84 5 9 Nurse Saul maturity strength relationships with datum temperature of 0 C for Mix #3 ................................ ................................ ................................ ................ 85 5 10 Nurse Saul maturity strength relationships with datum temperature of 10 C for Mix #3 ................................ ................................ ................................ ........... 85 5 11 Arrhenius maturity strength relationships with activation energy of 33500 J/mol for Mix #1 ................................ ................................ ................................ 87 5 12 Arrhenius maturity strength relationships with activat ion energy of 40000 J/mol for Mix #1 ................................ ................................ ................................ 88 5 13 Arrhenius maturity strength relationships with activation energy of 45000 J/mol for Mix #1 ................................ ................................ ................................ 88 5 14 Arrhenius maturity strength relationships with activation energy of 33500 J/mol for Mix #2 ................................ ................................ ................................ 89 5 15 Arrhenius maturity strength relationships with activation energy of 40000 J/m ol for Mix #2 ................................ ................................ ................................ 90 5 16 Arrhenius maturity strength relationships with activation energy of 45000 J/mol for Mix #2 ................................ ................................ ................................ 90 5 17 Arrheni us maturity strength relationships with activation energy of 33500 J/mol for Mix #3 ................................ ................................ ................................ .. 91 5 18 Arrhenius maturity strength relationships with activation energy of 40000 J/mol for Mix #3 ................................ ................................ ................................ 92

PAGE 14

14 5 19 Arrhenius maturity strength relationships with activation energy of 45000 J/mol for Mix #3 ................................ ................................ ................................ 92 5 20 Results of regression analys e s performed using MATLAB program ................. 95 5 21 Strength versus equivalent age plots for Mix #1 cured without Burlene covering at three different temperatures and standard condition ........................ 96 5 22 Strength versus equivalent age plots for Mix #1 cured with Burlene covering at three different temperatures and standard condition ................................ ...... 97 5 23 St rength versus equivalent age plots for Mix #2 containing appropriate and inappropriate amount of air ................................ ................................ ................. 99 5 24 Strength versus equivalent age plots for Mix #2 cured without Burlene covering at tw o different temperatures and modified standard condition .......... 100 5 25 Strength versus equivalent age plots for Mix #2 cured with Burlene covering at two different temperatures and modified standard condition ........................ 100 5 26 Strength versus equivalent age plots for Mix #3 cured without Burlene covering at three different temperatures and standard condition ...................... 102 5 27 Strength versus equivalent age plots for Mix #3 cured with Burlene covering at three different temperatures and standard condition ................................ .... 103 5 28 Plots of compressive strength at equivalent age of 8 hours versus slump of fresh concrete for Mix #1 ................................ ................................ .................. 107 5 29 Plots of compressive strength at equivalent age of 9 hours versus slump of fresh concrete for Mix #2 ................................ ................................ .................. 108 5 30 Plots of compressive strength at equivalent age of 8 hours versus slump of fresh concrete for Mix #3 ................................ ................................ .................. 108 5 31 Plots of comp ressive strength at equivalent age of 8 hours versus air content of fresh concrete for Mix #1 ................................ ................................ .............. 109 5 32 Plots of compressive strength at equivalent age of 9 hours versus air content of fresh con crete for Mix #2 ................................ ................................ .............. 110 5 33 Plots of compressive strength at equivalent age of 8 hours versus air content of fresh concrete for Mix #3 ................................ ................................ .............. 111 5 34 Plots of compressive strength at equivalent age of 8 hours versus unit weight of fresh concrete for Mix #1 ................................ ................................ .............. 112 5 35 Plots of compressive strength at equivalent age of 9 hours versus unit weight of fresh concrete for Mix #2 ................................ ................................ .............. 112

PAGE 15

15 5 36 Plots of compressive strength at equivalent age of 8 hours versus unit weight of fresh concrete for Mix #3 ................................ ................................ .............. 113 6 1 Locations and dates of the field studies performed. ................................ ......... 117 6 2 Installed temperature loggers at four different locations ................................ ... 119 6 3 Installed temperature logger s at the corner and edge of the slab ..................... 119 6 4 Curing of the specimens produced from the 5 th concrete ................................ 121 6 5 Drawing for the exact locations of the installed temperature loggers ................ 121 6 6 Compressive strength test for field generated maturity strength curve ............. 123 6 7 Recorded temperature histories for the specimens produced by different batches of concrete ................................ ................................ .......................... 125 6 8 Developed maturity strength cur ves from different batches of concrete before waiting time adjustment ................................ ................................ .................... 128 6 9 Comparisons of the temperature time plots for different locations of concrete slab and protection specimens ................................ ................................ ......... 130 6 1 0 Comparisons of the predicted strength time plots for different locations of concrete slab and protection specimens ................................ .......................... 131 6 1 1 Reco rded temperature histories for the specimens produced by different batches of concrete ................................ ................................ .......................... 133 6 1 2 Developed maturity strength s from different batches of concrete before waiting time adjustment ................................ ................................ .................... 136 6 1 3 Curing of the protection specimens and the last concrete slab ........................ 137 6 1 4 Comparisons of the temperature time plots for the co ncrete at different locations of slab and the protection specimens ................................ ............... 138 6 1 5 Comparisons of the predicted strength time plots for the concrete at different locations of slab and the protection specimens ................................ ............... 139 6 1 6 Comparisons of the different maturity strength curves and plots of protection specimens ................................ ................................ ................................ ........ 141 6 1 7 Preparation t ime frames for the different batches of concrete ......................... 142 6 18 Comparisons of developed maturity strength curves and plots obtained from both field studies after adjustment of different waiting tim es ........................... 143

PAGE 16

16 A 1 Temperature laboratory condition ................................ ................................ .......................... 151 A 2 Variations of temp erature measurements from same temperature sensors /loggers cured under ambient laboratory condition .............................. 151 A 3 Temperature laboratory co ndition ................................ ................................ .......................... 152 A 4 Variations of temperature measurements from same temperature sensors /loggers cured under ambient laboratory condition .............................. 152 A 5 Temperature F environment control chamber ................................ ................................ ........... 153 A 6 Variations of temperature measurements from same temperature sensors /loggers cu red in 113 F environment control chamber ......................... 153 A 7 Temperature curing tank ................................ ................................ ................................ ........ 154 A 8 Variations of temperature measurements from same temperature sensors /loggers cured in standard curing tank ................................ ................. 154 A 9 Equivalent ages from COMA meter and Intelli Rock maturity meter under three different curing conditions ................................ ................................ ........ 155

PAGE 17

17 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosop hy APPLICATION OF THE MATURITY METHOD FOR THE PREDICTION OF EARLY STRENGTH OF CONCRETE UNDER VARIOUS CURING CONDITIONS By Ohhoo n Kw on December 20 13 Chair: Mang Tia Co chair: Charles Gregola Major: Civil E ngineering Replacing concrete slabs on the highw ay facilities requires a contractor to close the lane until the replaced concrete slab ha s enough strength as a pavement In order to test the strength of the concrete in the replaced slab to determine the time to open to traffic, cylindrical specime ns cured under identical curing condition are made and tested However, the strength o f the concrete specimens does not accurately represent the actual strength of th e in place concrete. Furthermore, the compressive strength differences between different l ocations of the replaced concrete slab were founded to reach a s high as 3 0% at the same curing time For t hese reason s maturity method was proposed to determine the strength of replaced concrete slab as an alternative of testing s. In order to determine the most appropriate testing procedures to be used to obtain accurate strength prediction of the concrete used to slab replacement project, various possible factors which could affect the maturity strength prediction including mat urity systems, specimen size, fresh concrete properties, curing conditions, maturity functions, and modeling functions were evaluated throughout the laboratory experiment.

PAGE 18

18 Investigation was also made to evaluate t he effectiveness and relia bility of the ma turity strength prediction throughout multiple field studies. The laboratory generated maturity strength curve was compared to the one generated at the project site to validate the strength prediction of the maturity method. S trength prediction s for the di fferent locations of the replaced concrete slab were made to fin d the weakest location at the same curing time and thus the most appropriate location for installing temperature logger was determined. The results from this study shows that the maturity stre ngth prediction using the proposed testing procedure is a very reliable and convenient method for predicting early age strength of concrete used in slab replacement project

PAGE 19

19 CHAPTER 1 INTRODUCTION 1.1 Background Replacing concrete slabs on highway faci lities requires a contractor to close the lane, remove cracked and shattered concrete slabs, place new concrete in the repair area, and reopen the lane to traffic within a 10 to 12 hour window. To minimize the effect of the repair work, work is generally s tart ed at night time and done by next morning which includ es curing of the placed concrete. According to Section 353 of the FDOT Standard Specifications for Road and Bridge Construction, concrete specimens are required to have a 24 hour compressive streng th of 3,000 psi, and 2,200 psi prior to allowing traffic on the new pavement. In order to test the strength of the replaced slab, the specification requires making a minimum of four cylindrical specimens and curing the specimens by identical conditions use d in curing the replaced slab ( Freiesleben Hansen et al., 1997; Nixon et al., 2008 ). However, the testing procedure pos es the following challenges: Testing concrete cylinder requires a testing laboratory to be available at early morning hours on site, thus adding cost to the project. The measured strength of the concrete specimens may not represent the actual strength of the concrete in the replaced slab. Difference in compressive s trength as high as 20% ( Mohsen et al., 2004 ). For these reasons, the maturity method has been recommended to estimate strength of in place concrete as an alternative or verification method in most state DOTs including Florida DOT ( Bagheri Zadeh et al., 2007 ). Th e concept of concrete multiplied by the average temperature above freezing that a slab

PAGE 20

20 Based on this definition, he further developed the law for rela tionship between concrete approximately the same strength whatever combination of temperature and time goes to een done by other tool to predict concrete strength. 1.2 Problem Statement Although the maturity method is a proven method for indicating concrete strength for normal con crete placements and has the potential to be used to determine concrete strength for early opening to traffic, some limitations for the maturity method have been reported in previous research as follows: Different curing temperature can produce different M aturity Strength relationship and can result in the different prediction of long term strength of concrete ( Wade et al., 2006 ). In order to increase the accuracy of the maturity method, a sufficient amount of surface moisture is required for appropriate hy dration of concrete ( Tank et al. 1991 ). Fresh concrete properties such as entrained air, moisture content and unit weight affect the strength of the concrete and thus Maturity Strength relationship ( Carino et al., 1991 ). For these reasons, ASTM C1074 spec ifies that a new Maturity Strength relationship has to be developed for each different concrete mixture and sufficient amount of moisture must be supplied during curing process progressing. Also, a standard curing temperat ure, 73 F, is recommended as a reference temperature. However, the proposed limitations and recommendations by other studies are mostly based on long term maturity strength prediction for normal concrete Clear guidance

PAGE 21

21 and guidelines have not been established to use the maturi ty method to predict strength for high early strength concrete at the early age of 4 to 8 hours. 1.3 Hypothesis The hypotheses for this study are as follows: M aturity method using Arrhenius equation can be used to predict strength of the rep laced concrete slab to determine the time to open to traffic because : a) a t this age of concrete, cross over effect due to high early curing temperature may not have occurred. b) a t this age of concrete, improper hydration due to evaporation of all surface water may not have occurred. For accurate strength prediction, maturity strength curve should be generated by the same concrete that has the same mix design, fresh concrete properties and preparation time of the concrete to be used for strength predictio n In case that the variation of th e se factors over the tolerance ranges were detected, the developed maturity strength curve can be calibrated for the variation of these factors and thus, accurate strength prediction can be made. 1.4 Objectives The primar y objectives of this study are to evaluate effects of various factors on the strength maturity relationship and to determine if the maturity method can be used to determine early age strength of high early strength concrete to facilitate concrete slab repl acement. Detailed objectives are as follow: To evaluate effects of various factors on the compressive Maturity Strength relationship of concrete at early age. To develop appropriate test procedures for applying maturity method to predict early age strength of concrete. To validate the accuracy of the prediction of maturity method using the proposed test procedures. 1.5 Scope of Study This study investigates the effects of various curing conditions and variation of fresh concrete properties on the prediction of concrete strength at early age using the maturity method. As shown in Table 1 1, t wo curing variables (Curing Temperature and

PAGE 22

22 Curing Type) were used to simulate the various curing conditions for slab replacement projects in Florida. The four maturity s ystems considered are currently most widely used in Florida The effects of the fresh concrete properties that may have an effect on the application of maturity method were also investigated With the numerous variables considered, this research is aiming to develop an appropriate procedure for the use of maturity method for determining early strength of concrete in replaced slab s to determine time for opening to traffic. Table 1 1 Parameters considered in the study Curing Temp erature Curing Type Types of Maturity Systems Fresh Concrete Properties F F F t emperature Burlene C enter Rock meter concrete temperature c ontent w eight t ime 1.6 Research Approach and Methodology In achieving the set objectives in this study the following tasks were performed: 1.6.1 Literature Review A thorough review of past and curren t literature was conducted on the theory and practice of maturity method for concrete, state of the practice for Early Opening to Traffic (EOT) slab replacement, materials and construction specifications in use and guidelines for use of maturity method fo r slab replacement. 1.6.2 The First Set of Experiment s The main objective of the first experimental design was to evaluate the effectiveness of different temperature sensors/loggers under various curing environments and to select the most appropriate one t o be used in the maturity method

PAGE 23

23 for concrete. The differences between the results from the 6 12 cylinders and those from the cylinders were determined to find the most efficient specimen size for the second set of experiment s and field stud ies tha t were planned in this research. Two different concrete mix designs were used. The mix designs for these two concrete mixes represent typical concrete mixes which have been used in slab replacement application s in Florida. 1.6.3 The Second Set of Experime nt s T he second experimental design aim ed to evaluate the possible effects of different placement and curing environments on the predicted strength of concrete from the maturity method and to determine the most appropriate procedure to be used to obtain acc urate strength prediction of concrete. The maturity system and specimen size which were determined to be the most effective in the first set of experiment s were used and t hree different concrete mix designs were used. The mix designs for these three concre te mixes represent typical concrete mixes which have been used in slab replacement application in Florida. 1.6.4 Field Study The developed strength prediction procedure using the maturity method was applied to actual slab replacement project s in Florida to evaluate its effectiveness and reliability. The recorded temperature histories at different locations of actual slabs were used to calculate the equivalent age or the temperature time function (TTF) and the prediction of strength was made by a laborator y developed maturity strength for the concrete used in the actual slab. A total of two field studies were performed.

PAGE 24

24 Figure 1 1. Schematic diagram for the research approac h

PAGE 25

25 CHAPTER 2 LITERATURE REVIEW 2.1 Slab Replacement Project Slab replacement i s one of the typical method s used to repair severally deteriorated concrete slab s with relatively low project cost s than the other repair solutions such as overlay and reconstruction (Tia et al., 2008) According to the slab replacement guideline developed by California Department of Transportation (CADOT) in 2004, slab replacement need s to be applied for the following concrete slab conditions : When slabs have 2 or more corner breaks. When slabs have cracks into three or more pieces with interconnected crac ks developing between cracks or joints. When slabs have longitudinal or transverse cracks with 13 mm or more crack width When slabs have cracks having 150mm or more sp a lling and loss of concrete f ro m the crack centerline. When slabs have defects due to la ck of support such as settlement, base failure and excessive curing. In many h ighway agencies, to replace deteriorated concrete slabs, transportation is often restr icted, and thus this repair work s always aims to be done in very short time window Figu re 2 1 Process of slab replacement project

PAGE 26

26 Figure 2 1 shows the schematic proces s of a slab replacement project conducted by the Florid a Department of Trans portation According to the four phases in Figure 2 1 the phase whic h dictates project t ime window is the curing of the placed concrete. Therefore in most slab replacement high early strength conc rete mix designs which meet FDOT criteria for the replaced concrete slab requir ing to have more than 6 hour curing time and co mpressive strength of 2,200 psi prior to allowing traffic on the new pavement have been used. 2. 2 Strength Prediction with Maturity Method In 1949 McIntosh found that the rate of concrete strength gain is highly related with the curing temperature and a rate of hardening at any moment is directly proportional to the amount by which the curing temperature exceeds the datum With the concept of basic age used as a concrete hardening index, he defined that the trend of basic age versus compressive strength are very similar in the To predict strength of concrete Nurse introduced temperature time factor (TTF) in 1949 and Saul proposed a maturity strength prediction which is known as the Nurse Saul maturity method Figure 2 2. Diagram showing concept of the maturity method

PAGE 27

27 As shown in Figure 2 2 TTF can be calculated as a function of concrete temperature and time. Strength prediction can be made by applying calculated TTF to th e lab oratory developed maturity strength relationship with the notion that the same concrete mixtures will have the same strength at the same point of maturity. Though the maturity method has been proven to be relatively simple and accurate in the predict ion of concrete strength throughout various previous studies it is not widely applied in actual road way project s According to a previous study 29 states out of a considered 32 states did not have any guide lines for the application of maturity strength prediction (Tepke et al., 2007). Thus, for the last decade, strength prediction by maturity method has been recommended for use by the Federal H ighway Administration (FHWA) 2. 2 1 Nurse Saul M aturity F unction The Nurse Saul maturity function, which was or iginally proposed by Nurse and Saul in 1951, uses TTF as a maturity index. Because of its simplicity and fairly accurate strength prediction, many previous researchers recommended the use of this maturity function. The following equation shows the calculat ion of TTF using the concrete temperature history in t he Nurse Saul maturity function : (2 1) Where, M(t) = Maturity index ( F hours), or temperature time factor (TTF), = Time interval ( day s or hours), T a = T o = 2. 2 2 Arrhenius M aturity F unction The Arrhenius maturity function, which was developed by Freiesleben Hansen and Pedersen in 1977, is based on the rate of chemical reaction in concrete. Equivalent

PAGE 28

28 age is used in this maturity function as a maturity index. The following equation shows the calculation of equivalent age : (2 2) Where, t e = Equivalent age at a specified temperature, T s Q = Apparent activation energy, or activation energy divided by T a = Av = Time interval ( day s or hours), T s = 2. 3 Comparisons of Maturity Functions The TTF used in Nurse Saul maturity function as a maturity index can be transformed to a n equivalent age at a specified temperature. (Ca rino et al. 1983) As a result both Nurse Saul and Arrhenius maturity functions use the same maturity index and equivalent (2 3) (2 4) Where, = Age conversion factor Equation 2 4 shows the basic f o r m o f both maturity functions with the us e of an age conversion factor. Thus, it can be seen that both maturity functions use the same idea that for a concrete to reach the same strength under different curing temperatures, it mu st have the same equivalent age (Wade et al., 2006; Nixon et al., 2008). The only difference between these two maturity functions is the use of different age conver sion fac tors.

PAGE 29

29 Figure 2 3 Comparisons of age conversion factors for both m aturity functions Figure 2 3 shows relationships between age conversion factor and concrete curing temperature for both maturity functions The Nurse Saul maturity function uses an ag e conversion factor that has a linear relationship with concrete curing temperature. On the other hand, the Arrhenius maturity function uses a nonlinear relationship between age conversion factor and concrete curing temperature. Carino reported that the no nlinear age conversion factor used in the Arrhenius maturity function gives better fit to the nonlinear concrete hydration rate under different curing temperatures, and thus Arrhenius maturity function gives better prediction of concrete strength (Carino, 1991). On the other hand, other studies have recommended the consideration of both maturity functions because concrete hydration rate is mainly affected by mix design and some mix designs show better fit to the linear age conversion factor used in Nurse Sa ul maturity function under certain curing conditions. 0 0.5 1 1.5 2 2.5 3 3.5 4 0 10 20 30 40 50 60 Age Conversion Factor, Datum = Datum = AE = 33500J/mol AE = 40000J/mol AE = 45000J/mol

PAGE 30

30 2. 4 Limitations of Maturity M ethod According to the previous studies conducted by other researchers, some factors that affect concrete hydration such as curing temperature, curing humidity, and fresh concrete properties cause inaccurate maturity strength prediction. 2.4.1 Effect of Curing Temperature In 1962, Alexander and Taplin found that different curing temperatures have an effect on the maturity strength relationship. They developed maturity st rength curve s under different curing temperatures, 41 F (5 C) 70 F (21 C) and 108 F (42 C) f ro m a single mixture with Nurse Saul maturity function (Wade et al., 2006). Figure 2 4 Comparisons of maturity strength curves developed by Nurse Saul maturity function under different curing temperatures [ Adapted from Wade, S. S., Schindler S. K., Barnes, R W., and Nixon, J. M. 2006. Evaluation of the maturity method to estimate concrete strength. Research r eport for ALDOT, Contact N o. 930 590 (Page 21 Figure 2 13). Auburn University, Auburn, AL. ] Accordi ng to the maturity concept, the three developed three maturity strength curves sh ould have identical trend However each maturity strength curve had distinct ly 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 0 200 400 600 800 1000 Compressive Strength (psi) days)

PAGE 31

31 different trend s as shown in Figure 2 4 At early age, the specimens cured under lower temperatu re showed lower strength while the specimens cured under higher temperature showed higher strength at the same maturity. On the contrary, at late age, the specimens cured under higher temperature showed lower strength while the specimens cured under lower temperature showed higher strength at the same maturity In 19 68, Verbeck and Helmuth confirmed th is phenomenon and named it They found that the crossover effect is mainly affected by initial curing temperature T he strength of concret e exposed to the higher temperature at early age is greater than the strength of concrete exposed to the lower temperature and the TTF calculated by Nurse Saul maturity function cannot explain the different rate of chemical reaction of the concrete at diff erent curing temperature s In addition, impermeable hydration product around cement grains can be made due to the rapid hydration of concrete at the high curing temperature and which result s in long term strength loss. Verbeck and Helmuth concluded that ra pid hydration rate and long term strength loss under high curing temperature cause s the cro ssover effect (Wade et al., 2006). In 1991 Carino developed similar maturity strength curve s with the Arrhenius maturity function under three different curing temp eratures, 54 F ( 12 C), 70 F (21 C), and 90 F ( 32 C) f ro m a single concrete mixture ( Nixon et al., 200 8 ). Figure 2 5 shows the developed maturity strength curves in his study. It can be seen that at early age, the three curves show a relatively identical trend. However at the equivalent age of one day

PAGE 32

32 or later, significant strength loses on the concretes cured under higher curing temperature are detected. Figure 2 5 Comparisons of maturity strength curves developed by Arrhenius maturity function under different c uring temperatures [Adapted from Nixon, J. M., Schindler A. K., Barnes, R. W., and Wade, S. A. 2008. Evaluation of the maturity method to estimate concrete strength. Research r eport for ALDOT, Contact N o. 930 590 (Page 22, Figure 2 11). Auburn University, Auburn, AL. ] P revious studies conducted by other researchers found that neither Nurse Saul nor Arrhenius maturity function can perfectly account for the rate of strength development under different curing temperature (Guo, 1989; Tank et al., 1991, Wade e t al., 2006; Nixon et al., 2008). However, according to the result of Ca experiments, Arrhenius maturity function may give more accurate strength prediction s at early age. 2.4.2 Effect of Moist Curing In 1928, Gonnerman and Shuman conducted experime nt s to evaluate the effect of different moist curing time on the strength development of concrete. They generated strength versus time plots under same temperature but different moist conditions. 0 1000 2000 3000 4000 5000 6000 7000 0 5 10 15 20 25 30 Compressive Strength (psi) Equivalent Age at 23 (day) 12 C 21 C 32 C

PAGE 33

33 According to the result s of their study specimens cured und er moist condition for longer time ha d higher long term strength and the strength difference between moist cured specimens and air cured specimens was more than 3500 psi for one year strength. However, there was no strength difference observed between the specimens cured under different moist conditions at the early age (Gonnerman et al., 1928) The observed strength loss may occur due to insufficient moisture supply for proper c ement hydration and also can be observed on the maturity versus plots. In 1991 Carino found that similar strength loss occurred in the maturity strength relationship and concluded that Since these strength loss result in i naccurate maturity strength prediction ASTM C 1074 specification (2004) recommended a sufficient amount of moisture to be supplied for accurate in place concrete strength prediction. 2.4.3 Effect of F resh Concrete P roperties 2.4.3.1 Water to cement ratio ASTM C 1074 specified that the same concrete must be produced in the lab oratory in order to predict accurate in place concrete using the maturity method. However, it is hard to produce exactly the same concrete in the laboratory because actual moisture co ntent of the aggregates c an not be accurately accounted for during batch ing of concrete. This variation in moisture content affect s the water to cement ratio of the concrete, which in turn affect s the compressive strength of concrete

PAGE 34

34 F igure 2 6 Effec t of water cement ratio on the compressive strength [Adapted from Alawode, O. and Idowu, O. I. 2011. Effects of water cement ratios on the compressive strength and workability of concrete and lateritic concrete mixes. P acific j ourna l of s cien ce and t echno logy, Volume 12 2 (Page 102, Figure 2 ). ] It is well known that compressive strength in concrete mixtures decrease s with addition of more water. In 2011, Alawode and Idowu conducted experiments to determine the effect of water to cement ratio on compressive strength. They produced concrete mixtures having different water to cement ratios in the range of 0.55 to 0.8 by controlling the amount of water. They concluded that the mixture having higher water to cement ratio showed lower unit weight and higher slump value. As show n in Figure 2 6 com pressive strength increases by decreasing water to cement ratio. 2.4.3.2 Air content Air content in fresh concrete c an be varie d using different materials involving several chemical reactions ( Nixon et al., 2008 ). Table 2 1 shows the effect s of various factors on air content which in turn affect the compressive strength of the concrete. (Wilson et al., 2011). 7 9 11 13 15 17 19 21 5 10 15 20 25 30 Compressive Strength (N/mm2) Time (day) W/C=0.55 W/C=0.6 W/C=0.65 W/C=0.7 W/C=0.8

PAGE 35

35 Table 2 1. Effect of concrete materials and production practices on air entrainment Material/Practice Change Eff ect on Air Content Cement Increase in Cement Content Decrease Increase in Fineness Decrease Increase in Alkali Content Increase SCMs Fly Ash (With High Carbon) Significant Decrease Silica Fume Significant Decrease Slag with Increasing Fineness De crease Metakaolin No Change Aggregate Increase in Maximum Size Decrease Sand Content Increase Chemical Admixtures Water Reducers Increase Retarders Increase Accelerators No Change High Range Water Reducers Increase Water to Cement Ratio Increa se Water to Cement Ratio Increase Slump Increase in Slump up to 6 in Increase High Slump ( > 6 in) Decrease Low Slump ( < 3 in) Decrease Production Increased Mixer Capacity Increase Mixer Speeds to 20 rpm Increase Mixer Time Increase Transport a nd Delivery Transport Decrease Long Hauls Decrease Retemperirng Increase Placing and Finishing Belt Conveyors Decrease Pimping Significant D ecrease

PAGE 36

36 The effect of air content has already been shown by previous research F or every 1% increase in tota l volume of air, a 5% decrease in compressive strength can be expected (Mindess et al. 2003). Figure 2 7 shows the relationship between volume of air and 28 day compressive strength for concrete at three different cement content levels It can be seen tha t at all cement content, the compressive strength generally decreases with increasing volum e of air under the same slump value These strength variation s, due to the different fresh concrete properties such as moisture and entrained air content can also be ob served in the maturity strength curve s (Nixon et al., 2008). Hence, in order to predict accurately the compressive strength of concrete using the maturity method, the maturity strength curve must be developed using the same fr esh concrete properties a s the in place concrete. Figure 2 7 Relationship between air content and 28 day compressive strength [Adapted from Wade, S. S., Schindler S. K., Barnes, R. W., and Nixon, J. M. 2006. Evaluation of the maturity method to estimate concrete strength. Re search report for ALDOT, Contact no. 930 590 (Page 23, Figure 2 14). Auburn University, Auburn, AL.] 0 5 10 15 20 25 30 35 40 0 1 2 3 4 5 6 7 8 9 Compressive Strength(Mpa) Air Content (%) 613 lb/yd 519 lb/yd 425 lb/yd

PAGE 37

37 2. 5 Functions for Modeling Maturity Strength Relationships In order to predict accurate concrete strength using the maturity method, it is important to de velop appropriate maturity strength curve s that show an identical trend with the actual strength versus maturity index plots ASTM C 1074 states that each concrete mix design has a unique m aturity Strength relationship, and a common equation that can expla in all the unique maturity strength relationships is required for modeling the maturity strength relationships. In previous research work, several types of equations have been proposed by different researchers to model the maturity strength relationships. The most widely used equations are exponential, hyperbol ic, and logarithmic functions and will be presented in the following sections. 2. 5 .1 Modified Exponential Function In 1956, Nykanen proposed an exponential equation which mostly depend ed on the w/c ratio of the concrete (Nixon et al., 2008). T he equation did not fit the actual strength data very well. A s a result, Freiesleb en Hansen and Pederson proposed a modified exponential function in 1 9 84. Their modified exponential equation has shown good fit t o many maturity strength data sets especially at an early age (Carino, 1991). ASTM C 1074 recommends to use the modified exponential equation for modeling maturity strength relationship of concrete. Freiesleben Hansen and modified exponential equation is as follows: (2 5) Where, S = Compressive strength (psi), S u = Limiting compressive strength (psi), M = Maturity index ( C hours or hours), = Characteristic time co nstant (hours), and = Shape parameter

PAGE 38

38 2. 5 .2 Modified Hyperbolic Function The modified hyperbolic function was suggested by Carino in 1991. He introduced an offset of maturity, M 0 to account for the fact that strength development does not begin until t he maturity reaches a certain point. It has also been shown by other studies that the modified hyperbolic function fits well for various concrete mix designs. ASTM C 1074 recommends to use the modified hyperbolic function for model ing maturity strength rel ationship as well. The modified hyperbolic function is as follows: (2 6) Where, M 0 = Maturity when strength development is assumed to begin ( C hours or hours), and k = Rate constant (1 / [ C hours] or hours). 2. 5 3 Logarithmic Function Plowm an proposed the logarithmic function in 1956 for modeling maturity strength relationship. Because of the simplicity of this function, it is the most widely used for generating maturity strength curve s Illinoi s Department of Trans portation (IDOT) IM 383 an d Texas Department of Transportation ( T X DOT ) Tex 426 A guide line s recommend to use the logarithmic function to develop maturity strength relationship s (Nixon et al., 2008). However, some limitations of the logarithmic equation were discovered by previous studies, and thus this function was not recommended by ASTM C 1074 specification. The relationship predicts ever increasing strength with increasing maturity. The linear relationship is not valid at very early maturities.

PAGE 39

39 Only intermediate maturity values result in an approximately linear relationship between strength and the logarithm of maturity (Carino, 1991). The logarithmic function is as follows: (2 7) Where, a = Constant (psi), and b = Constant (psi / h ours or psi /[ C hours])

PAGE 40

40 CHAPTER 3 PREPARATION OF CONCRETE SPECIMENS 3.1 Introduction In order to develop appropriate testing protocol for the a pplication of maturity method to slab replacement project s more than 1 500 cylindrical specimens from three different mix designs and for various curing conditions were produced and tested. In this chapter, the aggregates used for the three mix designs wil l be described. Also, fresh concrete test s used to evaluate the characteristic of the concrete mixtures will be explained. 3.2 Concrete Mix D esigns Three different concrete mix designs were used for two different sets of laboratory expe riments. Two mix d esigns, namely Mix #1 and Mix # 2 were used to evaluate different maturity systems in the first set of experiment s The two mix designs used for the first set of experiment s and an additional mix design namely Mix #3 were used to evaluate the effect of d ifferent curing conditions on the maturity strength relationship in the second set of experiment s Table 3 1 Mix # 1 used in both first and second set of experiment s Name Product Name Quantity QPL # Cement Type I/II Cement 850 LB AASHITO Type I/II C oarse Aggregate #57 Stone 1,720 LB Fine Aggregate Silica Sand 983 LB Air Ent Admixture Darex AEA 1.0 OZ AASHITO M 154 AEA Type D Admixture WRDA 60 25.5 OZ AASHITO M 194 Type D Type F Admixture Adva 120 38.3 OZ AASHITO M 194 Type F Type E A dm ixture Daraccel 384.0 OZ AASHITO M 194 Type E Water Water 32.0 GA

PAGE 41

41 Table 3 2. Mix # 2 used in both first and second set of experiment s Name Product Name Quantity QPL # Cement Type I/II Cement 850 LB AASHITO Type I/II Coarse Aggregate #57 Stone 1,7 20 LB Fine Aggregate Silica Sand 983 LB Air Ent Admixture Darex AEA 1.0 OZ AASHITO M 154 AEA Type D Admixture WRDA 60 25.5 OZ AASHITO M 194 Type D Type F Admixture Adva 120 38.3 OZ AASHITO M 194 Type F Type E A dmixture Daraccel 384.0 OZ AASH ITO M 194 Type E Water Water 32.0 GA Table 3 3 Mix # 3 used in the second set of experiment s Name Product Name Quantity QPL # Cement Type I/II Cement 850 LB AASHITO Type I/II Coarse Aggregate #57 Stone 1,775 LB Fine Aggregate Silica Sand 999 LB Air Ent Admixture Darex AEA 6.8 OZ AASHITO M 154 AEA Type D Admixture WRDA 60 8.5 OZ AASHITO M 194 Type D Type F Admixture Adva 120 68.0 OZ AASHITO M 194 Type F Type E A dmixture Daraccel 382.5 OZ AASHITO M 194 Type E Water Water 32.0 GA All three concrete mix designs represent typical concrete mix designs which have been used in slab replacement application in Florida. Tables 3 1, 3 2, and 3 3 show the mix designs for Mixes #1, #2, and #3, respectively 3. 3 Preparation of Concrete Ingr edients 3.3.1 Cement In accordance with the mix design, two different Type I/II P ortland cement s f rom two cement manufacturers CEMEX a nd Suwannee American Cement were used to

PAGE 42

42 produce concrete mixtures. Table 3 4 shows the physical properties of both cemen t s as measured by the cement manufacturer s and their corresponding AASHTO/ASTM Type I/II cement specification limits. Table 3 4 Physical properties of the Type I/II cement CEMEX Suwannee Specification Limit Loss on Ignition 0. 5 % 2.4 3.0 % Autoclave Expansion 0.04% 0.06% 0.8 % Time of Setting (Initial) 1 72 min 104 min 60 min Time of Setting (Final) 2 60 min 215 min 600 min 3 day Compressive Strength 3127 psi 3889 psi 1450 psi 7 d ay Compressive Strength 4 892 psi 5084 psi 2 470 psi 3.3.2 Aggregate Silica sand and #57 Oolite limestone were used as fine and coarse aggregate s Both aggregates were obtained from the CEMEX batch plant in Gainesville where they produce slab replacement concrete The physical property tests were c onducted for each set of laboratory experiments. Table 3 5 and 3 6 show the test results for both the fine and coarse aggregate s Table 3 5. Physical properties of the fine aggregate The First S et of L aboratory E xperiment s The Second S et of L aboratory E xperiment s SSD Specific Gravity 2.615 2.478 Apparent Specific Gravity 2.623 2.488 Bulk Specific Gravity 2.610 2.471 Absorption 0.177 0.274

PAGE 43

43 Table 3 6 Physical properties of the coarse aggregate The First S et of L aboratory E xperiment s The Seco nd S et of L aboratory E xperiment s SSD Specific Gravity 2.42 3 2.42 2 Apparent Specific Gravity 2.59 2 2.578 Bulk Specific Gravity 2.3 17 2.32 3 Absorption 4.5 69 4.2 58 In order to measure the exact amounts of moisture content for both aggregate s the s ilic a sand was oven dried by placing the sand in an 110 C oven for over 24 hours. Also, the #57 stone was soaked in the water for 20 hours and then drained for 50 minutes to achieve s aturated surface dry ( SSD ) condition. 3.3.3 Admixtures Four d ifferent types of admixtures were used in each concrete mix design to achieve high early strength in this slab replace ment project. They are as follows: Air Entraining Admixture: It was used to stabilize micr oscopic air bubbles in concrete and thus durability of the con crete can be improved Type D Admixture: It was used to reduce water and retard the setting time. Thus, the right consistency and retard ing of setting time could be achieved. Type E Admixture: It was used to reduce water and accelerate the rate of concrete hydration. Thus, early strength development of concrete could be achieved. Type F Admixture: Similar to the Type D admixture, it was used to reduce water and greater amount of water can be reduced. Table 3 7 shows the product names and producers of the d ifferent admixtures used in the laboratory study. All four types of admixtures were diluted in the water used to produce concrete mixtures right before adding the water to other mix ingredients

PAGE 44

44 Table 3 7 Admixtures used in the laboratory experiments T ype Product Name Producer Air Entrained Admixture Darex AEA W.R Grace Co. Type D Admixture WRDA 60 W.R Grace Co. Type F Admixture ADVA 120 (Mix # 1 and # 3) ADVA 140M (Mix # 2) W.R Grace Co. Type E Admixture Daraccel W.R Grace Co. 3.4 Temperature Sen sor /Logger I nstallation As part of the first set of experiment s temperature sensors /loggers from Humboldt, Command Center and Intelli Rock maturity systems were compared to each other for their accuracy and efficiency. Since concrete cylinder s do not hav e the same temperature at different location s all temperature sensors /loggers were pre installed in the middle of both, and cylinder molds before concrete was placed. Figure 3 1. Pre installed maturity sensors /loggers in the cylind er molds As shown in Figure 3 1 the temperature sensors /loggers were firmly fixed in the middle of the molds, and thus would not be displa c ed during pouring and vibrating of concrete T herefore appropriate comparison s between the measurements made by di fferent sensors /loggers can be made in the first set of experiment s

PAGE 45

45 3. 5 Tests for Fresh and H ardened Concrete In accordance with ASTM specification s fresh and hardened concrete test s were performed for every concrete batch All test results were used to develop and compare among the various maturity strength curves developed under various curing conditions. 3. 5 .1 Time of Set In order to determine the overall time frame for making and testing concrete specimens in the la boratory initial and final settin g time tests were performed for all mix designs. The setting time test was conducted in accordance with ASTM C191 specification Mortar specimens were obtained by sieving the fresh concrete mixtures through a No. 4 sieve and penetration test were performed in a temperature controlled room maintaining temperature of 73 F (23 Figure 3 2. Humboldt A cme p enetrometer and morta r container (Photos courtesy of author, Ohhoon Kwon).

PAGE 46

46 Since the concrete mix designs used for this research were designed to get high early strength, penetration tests were performed every 5 minutes after an elapsed curing time of one hour. The results of the setting time test for all three mix designs were used to determine the time to remove the cylinder mold and to run the compressive strength test. Figure 3 2 shows the Humboldt Acme Penetr ometer and mortar container used in the setting time test in this research. 3. 5 .2 Slump Concrete slump tests were performed to determine the consistency of the fresh concrete mixture. The test s were performed for every separate batch in accordance to the A STM C143 specification. The slump test was performed as soon as concrete mixtures were produced because slump decrease s as time passes. The test results were used to control the quality of each fresh concrete mixture and to evaluate the effect of slump on the maturity strength relationship. 3. 5 .3 Air C ontent Concrete air c ontent test s were performed to determine the amount of air in the concrete and for every separate batch All test procedures followed the ASTM C173 specification A T ype B p ressure meter w as used to measure the volume of air. Since air con tent decreases with the passing of time, an air content test was performed as soon as slump test was finished. 3. 5 .4 Unit W eight Concrete unit weight tests were performed to determine the density of the f resh concrete. In accordance with the ASTM C 138 specification, all measurements were made by using type B pressure meter and calculated by the equation below :

PAGE 47

47 (2 4) Where, M c = Weight of the measure holding the concrete M m = Weight of the empty concrete measure V m = Volume of the measure (0.247 ft 3 ) The results of the unit weight test were reported in unit s of pcf and done 10 minutes after production of the fresh concrete mixture The test results were used to evaluate the effect of fresh concrete properties on the maturity strength relationship. 3. 5 .5 Temperature of Concrete Mixture In accordance with ASTM C1064, fresh concrete temperature was measured for each batch As h igh mixture temperature can result in strength loss in hardened concrete, the measurement of the fresh concrete temperature was used to check whether it was within the normal range. The temperatu re of fresh concrete was measured within 5 mi nutes after completing the concrete mixture. The results were reported with an accuracy of 1 F 3.5.6 Compressive Strength Though m aturity method can be applied for pr edicting different types of concrete streng ths only compressive strength tests were performed in this study a s FDOT ha s a single criterion of compressive strength to determine the time to open to traffic for a slab replacement project. As part of the second set of experiment s c ompressive strength tests were performed on all concrete mixes to develop and compare maturity strength curves under various curing conditions. All compressive strength procedures followed ASTM C39 standard Because of the characteristic of high early strength concrete used in this study, t he compressive strength tests were performed at early curing times Table 3 8 shows the elapsed curing times when the compressive strength tests were tested

PAGE 48

48 Table 3 8. Curing time s for testing compressive stre n gth Curing Temperature Co mpressive Strength Testing Time 4, 6, 8, 24, 168 h ours 3, 4, 6, 8, 24, 168 h ours 7, 10, 13, 24, 48, 168 h ours An average value of the three t est result s of the specimens was used for e ach testing time. When an individual strength has a significant level of deviation (more than 10% of the average strength), an average value of the other two test results was used. The two flat surfaces of each cylinder were evenly ground by using a diamon d wheel grinder.

PAGE 49

49 CHAPTER 4 THE FIRST SET OF LABORATORY EXPERIMENT S 4 .1 Laboratory Experiment Design 4 .1.1 Main Objective s As discussed previously, the main objective of the first set of laboratory experiment s is to evaluate the effectiveness of differen t maturity systems under various curing temperatures and to select the most appropriate one to be used in the maturity strength prediction for the rest of the study. The differences between the results from the 6 12 12 linders were determined to find the most effective specimen size to be used in the second set of experiment s and the field stud ies planned for this study 4.1.2 Maturity S ystems Since this study aims to propose the most appropriate maturity strength predi ction guideline for a slab replacement project, different maturity systems must also be evaluated. F our commonly used maturity systems were evaluated in this set of experiment s 4.1.2.1 Humboldt c oncrete m aturity m eter The Humboldt system 4101 concrete mat urity meter is one of the most commonly used multichannel maturity meters in the U.S. Figure 4 1 shows a picture of the connected to the maturity meter and RS 232C cables can be connec ted from the maturity meter to a computer to download the collected data. The Humboldt maturity meter records temperature every half an hour for the first two day s and then every hour for the rest of the time. Accuracy is 1 C and the range of temperature reading is 10 C

PAGE 50

50 to 90 C with 1 C of temperature resolution. The maturity meter meets ASTM C 1074 specification and can calculate both the Nurse Saul and the Arrhenius maturity functions. For the Nurse Saul maturity function, a datum temperature with a ran ge of 20 C to 60 C can be put into the meter. For the Arrhenius maturity function, a reference temperature with a range of 0 C to 40 C and activation energy with a range of 0 J/mol to 200 000 J/mol can be program m ed Figure 4 1 Humboldt System 4101 co ncrete maturity meter with thermocouple wires (Photos courtesy of author, Ohhoon Kwon). 4.1.2.2 Command Center maturity system The Command Center maturity system consists of self powered, self recorded temperature logger and manufacturer supplied softwar e. The recorded temperature history and the computed maturity index can be downloaded, whenever the data need to be used, by connecting the temperature logger to either a Command Center pocket computer or to any computer where the Command Center maturity s oftware has been installed. The Command Center maturity system generates the Nurse Saul maturity function from the collected temperature data and the input datum temperature.

PAGE 51

51 Accuracy of the temperature logger is 1 C with 10 C to 85 C of recording range and 0.5 C of temperature resolution. It can record temperature at specified time interval ranging from every 1 to 225 minute s which can be set through the Command Center software. In addition, the time of start of data recording can be preprogrammed into the temperature logger so that it is not required to activate temperature logger during placing concrete in the field. A picture of a Command Center temperature logger is shown in Figure 4 2 Figure 4 2 Command Center maturity temperature logger s (P hotos courtesy of author, Ohhoon Kwon). 4.1.2.3 Intelli Rock maturity system Similar to the Command Center matu rity system, the Intelli Rock maturity system consists of self powered and self recorded temperature logger a data reader and manufacturer supp lied software. Figure 4 3 is a picture of the temperature logger and data reader used for this system. The recorded temperature history and the maturity index can be downloaded by connecting the data reader to the temperature logger. Once the data are down loaded to the reader, the display panel shows all the information

PAGE 52

52 and useful plots of downloaded data. It can provide two maturity functions namely Nurse Saul and the Arrhenius, and several time intervals of temperature reading can be chosen by selecting from 22 different types of temperature loggers In addition, if other time intervals are required, the supplier can program the temperature loggers to meet the needs. The temperature logger has an accuracy of 1 C with a measurement range of 5 C to 85 C and 1 C of the temperature resolution. Any datum temperature for Nurse Saul maturity function and any activation energy for Arrhenius maturity function can be programmed into the temperature logger by connecting the data reader to a temperature logger Figure 4 3 Temperature logger and hand held reader of the Intelli Rock maturity s ystem (Photos courtesy of author, Ohhoon Kwon). 4.1.2.4 COMA meter The COMA (COncrete MAturity) meter is a convenient method for measuring the maturity of newly cast con crete by equivalent age according to the Arrhenius method.

PAGE 53

53 Figure 4 4 shows a picture of the COMA meters. The principle of operation of a COM A meter is described in the product information sheet as follows: A glass capillary contains a liquid for which th e rate of evaporation varies with temperature according to the Arrhenius equation, which is the same function that is used to determine maturity of concret e from the temperature history. Figure 4 4 COMA m eter (Photos courtesy of author, Ohhoon Kwon) The closed capillary is placed on a card with a scale indicating maturity in equivalent age at reference temperature of 20C. By reading the position of the liquid in the capillary on the scale, the equivalent age can be obtained. The measured equivalent age by COMA meter is the same as the calculated equivalent age with a fixed activation energy of 40 0000 J/mol. Currently, two types of COMA meter s are available for measuring equivalent age of concrete. COMA 5, which has a scale of 0 to 5 day s of equivale nt age, is recommended to be used in the high early strength concrete. COMA 14 which has a

PAGE 54

54 scale of 0 to 14 day s of equivalent age is recommended to be used in other types of concrete. 4 .1. 3 Curing Conditions The following t hree different curing condition s were used for the first set of experiment s : 73 F in curing tank (standard curing condition) 11 3 F in environment control chamber Ambient condition in the lab oratory Since ASTM specification C192 concrete specimens sh a ll be moist cured a t 73 F 3.5 F 73 F water curing environment was used as a standard curing condition. The 113 F curing chamber condition was designed to make an extreme ly hot curing condition to resemble the historically highest temperature record ed in Florida 109 F at Monticello in 1931. A n ambient curing room was also used to simulate the variation of air temperature in the field. 4 .1. 4 Concrete Specimens A total of two batches of concrete were produced and tested in this set of experiment s For each batch of concre te, the following co ncrete specimens were tested. Ten ( 10 ) d: a) o nly the ambient condition in the laboratory was used for curing. Each specimen was instrumented separately with one of the five different temper ature sensors/loggers b) t wo replicate specimens were used for each of the five types of maturity systems Twenty eight ( 28 ) : a) e ach specimen was instrumented separately with one of the five different temper ature sensors/loggers b) t he specimens instrumented with the first four sensors/loggers were placed in the three curing conditions c) t he specimens instrumented with the Coma meter were placed only in the last two curing conditions d) t wo replicate spec imens were used per condition

PAGE 55

55 Fifteen ( 15 ) : a) t hese specimens were cured in the curing tank at 73 F b) t hree specimens were tested for compressive strength at each of the following five c uring times: 4, 6, 8, 24, 168 h ours c) i t is to be noted that since high early strength concr etes are used in this study, 7 day strength was adequate for characterization of the ultim ate strength of the concrete. 7 day strength was chosen to be used inste ad of 28 day strength in order to shor ten the time of the experiment. 4 .2 Evaluation of the Effectiveness of Maturity Systems This section presents the test results and findings f rom the first set of experiment s which evaluated the effectiveness of differe nt maturity systems to be used in the maturity method for predicting concrete strength. Two different concrete mix designs were used in this set of experiment s Since the results from the two concrete mix designs gave similar trends and findings, only the results from the first concrete mix design are discussed in this chapter. The results from the second concrete mix design are presented in Appendix A 4.2. 1 Comparison of Recorded Temperature Histories The temperature data as recorded by the different sen sors /loggers from the same concrete mix design under the same curing condition were plotted and observed to see how they compared to one another. Figure 4 5 shows the comparison of temperature time plots of the concrete specimens cured under ambient labora tory condition as recorded by the Humboldt, Command Center and Intelli Rock maturity meters. The temperature time plot of the ambient laboratory environment during the testing period was also shown in this figure. It appears that the temperatures as record ed b y the Humboldt meter are 2 to 5 F higher than those recorded by the Command Center and Intelli Rock temperature loggers while the temperatures recorded by the Command Center and Intelli Rock temperature loggers were similar to one another.

PAGE 56

56 Figure 4 5 Temperature laboratory condition Figure 4 6 V ariations of temperature measurements from same temperature sensors /loggers cured under ambient laboratory condition 70 80 90 100 110 120 130 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt Room temp 0 1 2 3 4 5 6 7 8 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt

PAGE 57

57 Figure 4 6 shows var iations of temperature measurements from two identical temperature sensors /loggers It can be seen that the temperature data from the two Humboldt sensors had greater variations than those from the other temperature loggers Figure 4 7 shows the comparison of temperature time plots of the conc rete specimens cured in the 113 F environment control chamber as recorded by the different maturity meters. The temperature time plot of the environment control chamber during the testing period was also shown in this figure. Similarly, it appears that the temperatures as recorded b y the Humboldt meter are 2 to 4 F higher than those recorded by the Command Center and Intelli Rock temperature loggers Figure 4 7 Temperature red in 11 3 F environment control chamber Figure 4 8 shows variations of temperature measurements from two identical temperature sensors /loggers It can be seen that all types of temperature loggers show 90 100 110 120 130 140 150 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt Room temp

PAGE 58

58 a similar range of measurement error but the temper ature data from the two Humboldt sensors had slightly bigger variations than those from the other temperature loggers Figure 4 8 V ariations of temperature measurements from same temperature sensors /loggers cured in 113 F environment control chamber Figure 4 9 shows the comparison of temperature time plots of the concrete specimens in the standard curing tank as recorded by the different maturity meters. The temperature of the curing tank is also plotted on the figure. The compressive strength of the i n this figure. At equilibrium conditions, the temperatures recorded by the Humboldt meter were 2 to 6 F lower than the tank temperature, while the temperatures recorded by t he Intelli Rock meter were 2 F higher than the tank temperature, and the temperature recorded by the Command Center meter matched the curing tank temperature. 0 1 2 3 4 5 6 7 8 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt

PAGE 59

59 Figure 4 9 Tempera ture curing tank Figure 4 10 V ariations of temperature measurements from same temperature sensors /loggers cured in standard curing tank 70 72 74 76 78 80 82 84 86 88 90 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt Room temp 0 1 2 3 4 5 6 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt

PAGE 60

60 Figure 4 10 shows variations of temperature measuremen ts from two identical temperature sensors /loggers Similarly, it can be seen that the temperature data from the two Humboldt sensors once again had greater variations than those from the other temperature loggers Overall temper at ure measurements from I nte lli R ock and C ommand C enter temperature logger s show essentially identical measurement histories. On the o ther hand the Humboldt thermal couple wire s show 2 to 5 F higher temperature measurement s at the concrete temperature range of 90 to 120 F and 2 to 6 F lower temperature measurement s at the range of 85 to 70 F In addition C ommand C enter temperature loggers have the smallest deviation between their temperature measurements at the same condition whereas the Humboldt thermal coup le wires have the bigge st deviation between their temperature measurements. 4 .2. 2 Comparison of Accuracy of Temperature Reading The accuracy and response time to the temperature change of the Humboldt, Command C enter and Intelli Rock maturity sensors /loggers were evaluated by pl acing them in an ice water bath as shown in Figure 4 11 The ice water had an exact temperature of 32 F as measured by a calibrated mercury thermometer as shown in Figure 4 12.The i ce water temperature was measured by different maturity sensors /loggers every 5 minutes during a period of 60 minutes. Two replicate tests were performed for a more reliable ev aluation of the different maturity sensors /loggers

PAGE 61

61 Figure 4 11 Different temperature sensors /loggers placed in ice water Figure 4 12 Temperature reading on a m ercury thermometer in ice water

PAGE 62

62 Figure 4 13 Temperature time plots as recorde d by different maturity sen sors /loggers in ice water (replicate 1) Figure 4 14 Temperature time plots as recorded by different maturity sensors /loggers in ice water (replicate 2 ) 20 30 40 50 60 70 80 0 10 20 30 40 50 60 Elapsed Time (minute) Humboldt R1 Intelli rock R1 Command center R1 Mercury thermometer R1 20 30 40 50 60 70 80 0 10 20 30 40 50 60 Elapsed Time (minute) Humboldt R2 Intelli rock R2 Command center R2 Mercury thermometer R2

PAGE 63

63 Figure s 4 13 and 4 14 show comparisons of temperature time plots as re corded by the di fferent maturity sensors /loggers in ice water. It can be seen that the two Command Center and the two Intelli Rock temperature loggers gave readings of 31.1 F to 32 F while the two thermocouple wires from the Humboldt maturity system gave constant readings of 33.8 F and 35.6 F in the ice water bath Also, both Intelli Rock temperature logger s can be noted to have a delay time of 15 minutes before the constant final reading was obtained. The slower response time of the Intelli Rock temperatu re logger s has also been observed in previous t est. 4.2. 3 Comparison of Equivalent Ages Generated by COMA Meter and Intelli Rock Maturity Meter As explained earlier the COMA meter measures equivalent age with the liquid in the glass capillary where the r ate of evaporation varies in accordance with the concrete temperature Figure 4 15 shows the e quivalent age reading for the COMA meter. Figure 4 1 5 Equivalent age reading for COMA meter

PAGE 64

64 Since measurement of the equivalent age only requires a reading of the scale behind the glass capillary, the COMA meter is considered as the most convenient maturity system. However, its reading is initially calibrated for the use of a reference temperature of 20C and activation energy of 40000 J/mol and thus, it may not be used for different reference temperature s and activation energies In order to evaluate the accuracy of the COMA meter, the readings of the equivalent age made by the COMA meter were compare d to those fro m the Intelli R ock maturity system under thr ee different curing conditions. A reference temperature of 20C and activation energy of 40000 J/mol were also used for the Intelli R ock maturity system. The corresponding computed equivalent ages from the Intelli Rock system are plotted in Figure 4 16. Figure 4 16. E quivalent ages from COMA meter and Intelli Rock maturity meter under three different curing conditions 0 0.5 1 1.5 2 2.5 3 3.5 4 0 5 10 15 20 25 Equivalant Age (day) Elapsed Time (hour) COMA Ambient 4 in Intelli rock Ambient 4 in COMA Ambient 6 in Intelli rock Ambient 6 in COMA 113 Chamber 4 in Intelli rock 113 Chamber 4 in

PAGE 65

65 As can be observed in Figure 4 15, t he readings made by COMA meter underestimated at early age and overestimated at the curing time of one day Though the COMA meter is easy to use it also has limitations related with accuracy and application. 4.2. 4 Effects of Embedded Temperature Sensors /Loggers on the Strength of Concrete S pecimens An evaluation was made to determine how embedding di fferent temperature sensors /loggers in concrete specimens may affect the compressive strength of the which were embedded with various temperature sensors /loggers were tested for their compressive strengths after curing for 200 hours in the ambien t room and 200 hours in the 113 F curing chamber. Two replicate specimens for each condition were tested. Figure 4 17 C embedded maturity sensors / loggers 4000 4500 5000 5500 6000 6500 7000 7500 8000 8500 Compressive Strength (psi) COMA meter Humboldt Command Intelli rock

PAGE 66

66 Figure 4 17 presents the comparison of compressive strengths of the concrete specimens with different embedded maturity sensors /loggers at the two curing conditions. It can be seen that the specimens containing the Intelli Rock temperature logger s show approximately 20% lower strength than the other specimens for both curing conditions. The concrete specimens containing the Command Center temperature loggers and the Humboldt thermocouples show similar strength as the concrete specimens without tem perature sensors/loggers Figure 4 18 Rock temperature logger The significant effect of the Intelli Rock temperature logger on the compressive strength of the concrete specimen is possibly due to the relatively large size of the temperature logger Figure 4 1 8 Intelli Rock temperature logger. It can be seen that the temperature logger takes up a significant portion of the cross section of the s pecimen.

PAGE 67

67 4.2.5 Evaluation of Different Maturity Systems Table 4 1 presents a summary of the observed characteristics of the different maturity meters evaluated. The costs of the different maturity sensors /loggers are also given in the table. Based on the r esults of the evaluation, the Command Center maturity system was chosen to be used for the second set of experiment s and field studies. This decision was based on the accuracy, resolution, response time and convenience of use of this system. Table 4 1 S u mmary of the observed characteristics of the different maturity meters evaluated Intelli Rock Command Center Humboldt COMA M eter Accuracy in Temp erature Reading No detectable error No detectable error Error of 2 N/A Frequency of Temperature Reading Pre determined by type of temp. logger (1 to 1440 minute) Determined by user programming (1 to 225 minutes) Every 30minutes for 2 day s and every 1 hour for remaining time N/A Duration of Temperature Reading Up to 365 day s depending on type of temp. logger 7 day s with 5 minutes frequency 28 day s with 15 minutes frequency Up to 334 day s Until the equivalent age reaches 14 day s Resolution of Temperature Re ading N/A Cost of Sensor / L ogger $ 33/each temperature logger $28/each temperature logger $0.81/foot of wire $28/each Convenience to U se Easy to use. Relatively big temperature logger may have negative effect on the structure strength Convenient and easy to use. Easy to use. Sensor need to be connected to meter continuously Convenient and easy to use. Gives only equivalent age

PAGE 68

68 4.3 Evaluation of the Effectiveness of Different Specimen S ize s pecimens Temperature sensors /logge rs from the Humboldt, Command Center and Intelli Rock maturity meters were also pre before concrete placement Only the ambient condition was used for both sizes of specimens to compare their temperature his tories. Figure s 4 19 4 20 and 4 21 show the comparisons of temperature histories of two different size s of specimens recorded by Intelli Rock temperature logger, Command C enter temperature logger and Humboldt maturity system respectively. Figure 4 19 Temperature histories of concrete specimens cured under ambient temperature as recorded by Intelli Rock temperature logger 70 80 90 100 110 120 130 140 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock D4 Intelli Rock D6 Room temp

PAGE 69

69 Figure 4 20 Temperature histories concrete specimens cured under ambient temperature as recorded by Command C enter temperature logger Figure 4 21 Temperature histories concrete specimens cured under ambient temperature as recorded by Humboldt maturity system 70 80 90 100 110 120 130 140 0 5 10 15 20 25 30 Elapsed Time (hour) Command Center D4 Command Center D6 Room temp 70 80 90 100 110 120 130 140 0 5 10 15 20 25 30 Elapsed Time (hour) Humboldt D4 Humboldt D6 Room temp

PAGE 70

70 In all three figures, it appea red that the temperature hi stories of specimens were 0 to 12 F higher than those specimens. On the other hand, specimens reached the high e st temperature approximately 1 hour earlier tha n the specimens. 4.3. 2 Comparison of S trength Development S pecimens To generate and compare the maturity stre ng th curves developed from the specimens, c ompressive stre ngth tests were performed for both size s of sp ecimens. The compressive stre n gth of three were tested at 4, 6, 8, 24, 168 and 336 hours On the other hand, the compressive stre ng th of one 6 specimens was tested at 5 hour s 40 minute s 7 hour s 40 minute s and 23 hour s 40 minute s Table 4 2 and Fig ure 4 22 show the result s of the compressive stre ngth tests for both and specimens. Table 4 2 Compresive str ength test results for Time (hour) 1 st Strength (psi) 2 nd Strength (psi) 3 rd Strength (psi) Ave Strength (psi) Time (hour) 1 st Strength (psi) 4 287 295 306 296 N/A N/A 6 1496 1529 1506 1510 5.67 1589 8 2549 2511 2527 2529 7.67 2640 24 5736 5841 5707 576 23.67 5979 168 7903 7735 N/A 7819 N/A N/A 336 8012 7699 N/A 7856 N/A N/A

PAGE 71

71 Figure 4 22 Comparisons of compressive strength time plots for specimens cured under ambient laboratory condition In Figure 4 22 it can be observed that the specimens have higher compressive stre ngths than those of at the same curing time. It can be e specimens causes higher rate of hydration of the concrete used and result s in higher stre ngth at the same curing time. 4.3. 3 E ffect of Specimen Size on the Maturity Strength R elationship As shown in the previous section different size s of specimens produce different temperature histories and different compressive strength for the same curing time. In order to evaluate the effect of different specimen size s on the maturity strength relationship, tw o maturity strength curves were generated. Nurse Saul maturity function was used to calculate TTF with datum temperature of 14 F and a modified exponential modeling function was used to generate the maturity strength c urve. There was no 0 800 1600 2400 3200 4000 4800 5600 6400 0 5 10 15 20 25 Compressive Strength (psi) Elapsed Time (hour) 4"x 8" specimen 6"x 12" specimen

PAGE 72

72 remarkable strength increase observed at the 14 day strength as compar ed to the 7 day strength which ha d a compressive strength of 7800 psi. Figure 4 23 M aturity strength plots for specimens cured under ambient laboratory condition Figure 4 23 shows the matu rity stre ng th plots for specimens and specimens As shown in Figure 4 23 there were no visible differen ces detected between the maturity stre ng th plots for and matu rity stren g th plots for specimens Th ere fore it can be concluded that the different specimen size s do not have any effect on the maturity stre ngth relantionship. For the co nvenience of use of the specimens, this specimen size was chosen to be used for the second set of experiment s and field studies 0 1000 2000 3000 4000 5000 6000 7000 100 300 500 700 900 1100 1300 1500 1700 Compressive Strength (psi) TTF ( hours, datum = 4in Actual Data 6in Actual Data

PAGE 73

73 CHAPTER 5 THE SECOND SET OF LABORATORY EXPERIMENT S 5.1 Laboratory Experiment Design 5.1.1 Main Objectives The main obje ctives of the second set of experiment s are to evaluate the possible effects of different placement and curing environments on the predicted strength of concrete from the maturity method and to determine the most appropriate procedure to be used to obtain accurate predicted strength of concrete. 5.1.2 Maturity System s Command Center maturity system was chosen to be used for the second set of experiment s because the maturity system was determined to be the most effective system in the first set of experiment s This decision was made based on the accuracy, resolution, response time and convenience of use of this system in the first set of experiment s 5.1.3 Curing Conditions The following f ive different curing t emperatures were used for the second set of expe riment s : 73 F in curing tank (standard curing condition) 7 3 F in environment control chamber 113 F in environment control chamber 4 3 F in environment control chamber Ambient condition in laboratory 73 F curing chamber conditions were used as a standar d curing temperature 113 F and 45 F curing chamber conditions were designed to simulate extremely hot and

PAGE 74

74 cold curing condition with consideration of the historically highest and lowest temperature record s in Florida of 109 F and 43 F respectively ( Wikipe dia 2013 ). An ambient room curing temperature was also used to represent the usual variation of air temperature in the field. For each curing temperature with the except ion of the standard curing condition, t he following t wo exposure conditions were app lied in the second set of experiment s Exposed to air Covered with Burlene curing blanket T hus a total of nine curing conditions including the standard curing condition were used for this set of experiment s 5.1.4 Concrete S pecimens Three different mi x designs were used for this set of experiment s and the following concrete spe cimens were tested for each mix design. Three replicate cylindrical specimens were tested at each of the five curing times (4, 6, 8, 24 and 168 hours) for specimens cured u nder the following five conditions: a) standard curing tank, b) 73 F and exposed to air, c) 73 F and covered with Burlene, d) ambient temperature and exposed to air, and e) ambient temperature and covered with Burlene. Three replicate cylindrical specimens were tested at each of the s ix curing times (3, 4, 6, 8, 2 4, and 168 hours ) for specimens cured under following two conditions: a ) 11 3 F and exposed to air, b ) 11 3 F and covered with Burlene Three replicate cylindrical specimens were tested at each of the s ix curing times ( 7, 10, 13, 24, 48, and 168 hours ) for specimens cured under following two conditions: a ) 4 3 F and exposed to air, b ) 4 3 F and covered with Burlene 18 cylindrical specimens to be instrumented for temperature reading : a) n ine curing conditions b) t wo replicate specimens per each cu ring condition

PAGE 75

75 Each batch was divided into 4 small batches due to lack of time for the hardened concrete tests. Also two replicate batches were made for each of the three concrete mix designs. 5.2 Fresh Concrete Properties of the Second Set of E xperiment s The fresh concrete properties of the eight replicate batches of Mix # 1, # 2, and # 3 are presented in Tables 5 1, 5 2, and 5 3 respectively. Though the eight replicate batches of each concrete mix design were meant to be identical, there were some variat ions in their fresh concrete properties as can be observed from the test value s given in these three tables. Table 5 1 Fresh c oncrete p roperties of eight batches of concrete Mix # 1 Mix # 1 Curing C ondition Ambient Hot Curing Chamber Standard Curing Cha mber Cold Curing Chamber 71 73 111 114 72 74 44 55 Curing Humidity (%) 40 55 35 45 45 70 60 90 Mix ture P roperties Replicate No. (Mix No.) 1 (3) 2 (3) 1 (9) 2 (10) 1 (13) 2 (14) 1 (19) 2 (20) Coarse A ggregate Moisture C ontent (%) 6.15 7.55 6.74 6.86 6.57 5.95 5 .67 5.94 Slump (in) 9.00 6.25 6.25 8.25 6.75 6.50 7.00 7.25 Mix T 85 85 85 86 76 77 76 76 Air C ontent (%) 1.90 2.00 2.00 1.90 2.20 2.30 2.10 2.00 Unit W eight (pcf) 14 9.0 148. 6 14 9.0 146.5 146.1 145.7 147. 4 147. 8 Mixing R oom T emperature 83 85 81 82 77 79 75 73 Elapsed T ime to R emove C ylinder M old (hour) 3 3 2 2 3 3 6 6

PAGE 76

76 Table 5 2 Fresh c oncrete p roperties of eight batches of concrete Mix # 2 Mix # 2 Curing C ondition Ambient Hot Curing Chamber Standard Curing Chamber Cold Curin g Chamber 71 73 111 114 72 74 44 50 Curing Humidity (%) 40 55 35 45 45 70 60 90 Mix ture P roperties Replicate No. (Mix No.) 1(1) 2(2) 1(11) 2 (12) 1(15) 2(16) 1(21) 2(22) Coarse A ggregate Moisture C ontent (%) 5.46 6.00 6.29 6.33 6.53 6 .66 7.34 6.16 Slump (in) 6.75 4.25 4.00 4.00 4.00 4.25 4.25 5.50 Mix T 83 82 82 82 78 75 79 79 Air C ontent (%) 5.10 4.10 4.20 4.90 7.20 7.60 7.10 7.30 Unit W eight (pcf) 139.7 143.3 143.3 141.7 138. 1 137. 7 139.3 138. 5 Mixing R oom T empera 82 81 81 81 75 72 75 75 Elapsed T ime to R emove C ylinder M old (hour) 3 3 2 2 3 3 6 6 Table 5 3 F resh c oncrete p roperties of eight batches of concrete Mix # 3 Mix # 3 Curing Condition Ambient Hot Curing Chamber Standard Curing Chamber Cold Cur ing Chamber Curing 71 73 111 114 72 74 44 50 Room Humidity (%) 40 55 35 45 45 70 60 90 Mix ture P roperties Replicate No. (Mix No.) 1(5) 2(6) 1(7) 2(8) 1(17) 2(18) 1(23) 2(24) Coarse A ggregate Moisture C ontent (%) 5.81 6.64 6.81 7.20 6.03 6.53 7.18 7.23 Slump (in) 8.00 8.25 7.50 6.75 8.25 8.00 8.00 8.00 Mix T 84 84 79 77 76 79 77 77 Air C ontent (%) 2.50 2.80 2.00 2.10 2.70 2.50 2.10 2.40 Unit W eight (pcf) 145. 3 143.8 146.5 147. 4 144.5 145.7 147. 4 146.9 Mixing R oom T emperatur 82 82 76 72 68 75 75 73 Elapsed T ime to R emove C ylinder M old (hour) 3 3 2 2 3 3 6 6

PAGE 77

77 One possible reason for the variations in air content from one replicate batch to another is the change in mixing temperatur e, especially in Mix # 2. Figure 5 1 sho ws a plot of air content versus m ixing room temperature for Mix # 2. It can be observed that air content generally increased with decreasing room temperature. Figure 5 1. Air content versus mixing room temperature for Mix # 2 Another source of variatio n in fresh concrete properties is the variation in moisture content of the aggregates, which may not be accurately accounted for during batching of materials for the con crete mixtures An underestimation of the actual moisture content of the aggregates may result in too much mixing water added, which can cause a higher slump and higher water cement ratio of the concrete. Adding too much water may also cause a reduction in unit weight and air content of the fresh concrete (Wilson et al., 2011). 2 3 4 5 6 7 8 9 10 68 70 72 74 76 78 80 82 84 86 Air Comtents (%) Air contents vs. Mixing temp. Linear (Air contents vs. Mixing temp.)

PAGE 78

78 5.3 Evaluat ion of Maturity Functions with Test Result s 5.3.1 Introduction B oth Nurse Saul and Arrhenius maturity functions were originated from same concept that for a concrete to reach the same strength under different curing temperatures, it must have the same equi valent age. (Wade et al., 2006). Since its non linear age conversion factor represents strength development of concrete well, the Arrhenius maturity function has been reported for giving more accurate predictions of concrete strength. However, some studies have reported that, for certain mix design s and curing conditions, the Nurse Saul maturity function gives better prediction than the Arrhenius maturity function, and thus, some researchers have recommended for both maturity functions to be evaluated with regards to their accuracy in predicting strength of concrete in practice (Wade et al., 2006). T his chapter presents the results of an in depth evaluation of different maturity functions for prediction of early strength of concrete for application in slab r eplacement. 5.3.2 Parametric Study on Nurse Saul Maturity Function For use of the Nurse Saul maturity function, different datum temperature s cause a change in age conversion factor and a change in the calculated TTF. Thus, using an appropriate datum tempe rature is very important for accurate prediction of concrete strength in the Nurse Saul maturity method. Many researchers proposed to use different datum temperatures (Wade, 2005) and most of them are in the range from 11 C to were applied to the test results from the second set of experiment s to develop maturity strength relationship under different curing temperatures.

PAGE 79

79 Since the differences in fresh concrete properties could affect the maturity strength relationship, batches of concrete that had the same or very similar fresh concrete properties were selected for use in the parametric stud y Table 5 4 shows the selected batches for each mix design Table 5 4 Selected concrete batches for the parametric study Mix #1 Curing Condition Med Temperature Hot Temperature Cold Temperature Mixture Identification Ambient R2 Slump (in) 6.25 6.25 7.00 Mix T emperature ( ) 85 85 76 Air C ontent (5) 2.0 2.0 2.1 Unit W eight (pcf) 148. 6 148. 6 147. 4 Mix #2 Curing Condition Med Temperature Hot Temperature N/A Mixture Identification Ambient R2 Slump (in) 4.25 4.00 Mix T emperature ( ) 82 82 Air C ontent (%) 4.1 4.2 Unit W eight (pcf) 143.3 143.3 Mix #3 Curing Condition Med Temperature Hot Temperature Cold Temperature Mixture Identification Ambient R1 Slump (in) 8.00 7.50 8. 00 Mix T emperature ( ) 84 77 77 Air C ontent (%) 2.5 2.1 2.4 Unit W eight (pcf) 145. 3 147. 4 146.9

PAGE 80

80 Figures 5 2 through 5 4 show various Nurse Saul maturity strength relationships developed with the data from the specimens cured under different curing te mperatures for Mix # 1. It can be observed that the datum temperature of 5 C gives the best matching between the maturity strength relationships developed from the specimens cured under three different curing temperatures. However, more than 25% of d ifferen ces in the values of TTF were observed between the specimens cured under three different temperatures at the critical strength of 2200 psi for both exposed curing and Burlene covering conditions. Figure 5 2 Nurse Saul maturity streng th relationship s with datum temperature of 5 C for Mix # 1 A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering. 0 500 1000 1500 2000 2500 3000 3500 4000 50 150 250 350 450 Compressive Strength (psi) TTF ( F hours) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 50 150 250 350 450 Compressive Strength (psi) TTF ( F hours) Med Temp High Temp Low Temp A B

PAGE 81

81 Figure 5 3 Nurse Saul maturity streng th relationships with datum temperature of 0 C for Mix # 1 A) S pecimens cured under exposed curing cond ition B) S pecimens cured with Burlene covering. Figure 5 4 Nurse Saul maturity streng th relationships with datum temperature of 10 C for Mix # 1 A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering. 0 500 1000 1500 2000 2500 3000 3500 4000 100 200 300 400 500 Compressive Strength (psi) TTF ( F hours) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 100 200 300 400 500 Compressive Strength (psi) TTF ( F hours) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 200 300 400 500 600 Compressive Strength (psi) TTF ( F hours) Mix #1 Exposed (Datum = Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 200 300 400 500 600 Compressive Strength (psi) TTF ( F hours) Mix #1 Burlene (Datum = Med Temp High Temp Low Temp A B A B

PAGE 82

82 Figures 5 5 through 5 7 show various Nurse Saul maturity strength relationships developed from the specimens cured under two different curing temperatures for Mix # 2 Similarly, it can be observed that the datum temperature of 5 C gives the best matching between t he maturity strength relationships developed from the specimens cured under two different curing temperatures. However, more than 15 % of d ifferences in the values of TTF were observed between the specimens cured under two different curing temperatures at t he critical strength of 2200 psi for both exposed curing and Burlene covering conditions. Figure 5 5 Nurse Saul maturity streng th relationships with datum temperature of 5 C for Mix # 2. A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering. 0 500 1000 1500 2000 2500 3000 3500 4000 100 300 500 700 Compressive Strength (psi) TTF ( F hours) Med Temp High Temp 0 500 1000 1500 2000 2500 3000 3500 4000 100 300 500 700 Compressive Strength (psi) TTF ( F hours) Med Temp High Temp A B

PAGE 83

83 Figure 5 6 Nurse Saul maturity streng th relationships with datum temperature of 0 C for Mix # 2. A) S pecimens cured under exposed curing cond ition B) S pecimens cured with Burlene covering. Figure 5 7 Nurse Saul maturity streng th relationships with datum temperature of 10 C for Mix # 2. A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering. 0 500 1000 1500 2000 2500 3000 3500 4000 100 250 400 550 Compressive Strength (psi) TTF ( F hours) Med Temp High Temp 0 500 1000 1500 2000 2500 3000 3500 4000 100 250 400 550 Compressive Strength (psi) TTF ( F hours) Med Temp High Temp 0 500 1000 1500 2000 2500 3000 3500 4000 200 300 400 500 600 Compressive Strength (psi) TTF ( F hours) Mix #2 Exposed (Datum = Med Temp High Temp 0 500 1000 1500 2000 2500 3000 3500 4000 200 300 400 500 600 Compressive Strength (psi) TTF ( F hours) Mix #2 Burlene (Datum = Med Temp High Temp A B A B

PAGE 84

84 Figures 5 8 through 5 10 show various Nurse Saul maturity strength relationships developed from the specimens cured under three different curing temperatures for Mix # 3. It best matching between the matu rity strength relationships developed from the specimens cured under three different curing temperatures. However, more than 3 0% of differences in the calculation of TTF were also observed between the specimens under three different curing temperatures at the critical strength of 2200 psi for both exposed curing and Burlene covering conditions. Figure 5 8 Nurse Saul maturity streng th relationshi ps with datum temperature of 5 C for Mix # 3 A) S pecimens cured under exposed curing condition B) S pecime ns cured with Burlene covering. 0 500 1000 1500 2000 2500 3000 3500 4000 50 150 250 350 450 Compressive Strength (psi) TTF ( F hours) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 50 150 250 350 450 Compressive Strength (psi) TTF ( F hours) Med Temp High Temp Low Temp A B

PAGE 85

85 Figure 5 9 Nurse Saul maturity streng th relationships with datum temperature of 0 C for Mix # 3. A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering. Figure 5 10 Nurse S aul maturity streng th relationships with datum temperature of 10 C for Mix # 3. A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering. 0 500 1000 1500 2000 2500 3000 3500 4000 100 200 300 400 500 Compressive Strength (psi) TTF ( F hours) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 100 200 300 400 500 Compressive Strength (psi) TTF ( F hours) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 200 300 400 500 600 Compressive Strength (psi) TTF ( F hours) Mix #3 Exposed (Datum = Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 200 300 400 500 600 Compressive Strength (psi) TTF ( F hours) Mix3 Burlene (Datum = Med Temp High Temp Low Temp A B A B

PAGE 86

86 For all different mix designs and datum temperatures, significant error s bet ween the Nurse Saul maturity strength relationships developed with the concrete cured under three different curing temperatures were observed. Thus, the use of the maturity strength relationship generated by Nurse Saul maturity func tion for prediction of s trength of concrete at early age under various curing conditions could result in substantial errors. 5.3.3 Parametric Study on Arrhenius Maturity Function Similar to the Nurse Saul maturity function, a change in activation energy directly affects the calc ulation of equivalent age and thus, using an appropriate activation energy is very important for accurate prediction of concrete strength with the Arrhenius maturity method. Many researchers have proposed and used various activation energies in their rese arch (Freiesleben Hansen et al., 1977; Carino, 1991; Tank et al., 1991) Samuel reported that an activation energy of 33500 J/mol showed accurate strength prediction for warm weather concrete curing, and an activation energy of 40000 J/mol showed accurate strength prediction for cold weather concrete curing. ASTM C 1074 suggests to use an activation energ y in the range of 40000 to 45000 J/mol for concrete made with Type I cement when no chemical admixture is used. Also, Freiesleben and Pedersen suggested an equation for calculating appropriate activation energy depending on the temperature of the concrete as follows: ( 5 1 ) Where, T a =

PAGE 87

87 For this parametric analysis three activation energies, namely (1) 33500 J/mol ( Freiesleben Hansen et al. 1977), (2) 40000 J/mol, and (3) 45000 J/mol (Wade et al. 2006 and ASTM C 1074, 2004) were used and applied to the test results of the second set of experiment s to develop maturity strength relationship under diffe rent curi ng conditions. In order to compare the maturity strength relationships generated by both maturity functions, t he same batches of concretes used for the parametric analysis of Nurse Saul maturity function were also used. Figure 5 11. Arrheni us maturity streng th relationships with activation energy of 33500 J/mol for Mix #1. A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering. 0 500 1000 1500 2000 2500 3000 3500 4000 2 5 8 11 14 Compressive Strength (psi) Equivalent age (hour) Mix #1 Exposed (AE = 33500 J/mol) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 2 5 8 11 14 Compressive Strength (psi) Equivalent age (hour) Mix #1 Burlene (AE = 33500 J/mol) Med Temp High Temp Low Temp A B

PAGE 88

88 Figure 5 12. Arrhenius maturity streng th relationships with activation energy of 40000 J/mol for Mix #1. A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering. Figure 5 13. Arrhenius maturity streng th relationships with activation energy of 45000 J/mol for Mix #1. A) S pecimens cure d under exposed curing condition B) S pecimens cured with Burlene covering. 0 500 1000 1500 2000 2500 3000 3500 4000 2 5 8 11 14 Compressive Strength (psi) Equivalent age (hour) Mix #1 Exposed (AE = 40000 J/mol) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 2 5 8 11 14 Compressive Strength (psi) Equivalent age (hour) Mix #1 Burlene (AE = 40000 J/mol) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 2 5 8 11 14 Compressive Strength (psi) Equivalent age (hour) Mix #1 Exposed (AE = 45000 J/mol) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 2 5 8 11 14 Compressive Strength (psi) Equivalent age (hour) Mix #1 Burlene (AE = 45000 J/mol) Med Temp High Temp Low Temp A B A B

PAGE 89

89 Figures 5 11 through 5 13 show various Arrhenius maturity strength relationships developed from the specimens cured under three different curing temperatures for Mix #1. For both exposed curing and Burlene covering conditions, the use of activation energy of 33500 J/mol gives the best matching between the maturity strength relationships generated from the specimens cured under three different curing temperatures. Also, the differen ces between the calculated equivalent ages at the critical strength of 2200 psi were less than 6% for both exposed curing and Burlene covering conditions. Figure 5 14. Arrhenius maturity streng th relationships with activation energy of 33500 J/mol f or Mix #2. A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering. 0 500 1000 1500 2000 2500 3000 3500 4000 2 7 12 17 Compressive Strength (psi) Equivalent age (hour) Mix #2 Exposed (AE = 33500 J/mol) Med Temp High Temp 0 500 1000 1500 2000 2500 3000 3500 4000 2 7 12 17 Compressive Strength (psi) Equivalent age (hour) Mix #2 Burlene (AE = 33500 J/mol) Med Temp High Temp A B

PAGE 90

90 Figure 5 15. Arrhenius maturity streng th relationships with activation energy of 40000 J/mol for Mix #2. A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering. Figure 5 16. Arrhenius maturity streng th relationships with activation energy of 45000 J/mol for Mix #2. A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering 0 500 1000 1500 2000 2500 3000 3500 4000 2 7 12 17 Compressive Strength (psi) Equivalent age (hour) Mix #2 Exposed (AE = 40000 J/mol) Med Temp High Temp 0 500 1000 1500 2000 2500 3000 3500 4000 2 7 12 17 Compressive Strength (psi) Equivalnet age (hour) Mix #2 Burlene (AE = 40000 J/mol) Med Temp High Temp 0 500 1000 1500 2000 2500 3000 3500 4000 2 7 12 17 Compressive Strength (psi) Equivalent age (hour) Mix #2 Exposed (AE = 45000 J/mol) Med Temp High Temp 0 500 1000 1500 2000 2500 3000 3500 4000 2 7 12 17 Compressive Strength (psi) Equivalent age (hour) Mix #2 Burlene (AE = 45000 J/mol) Med Temp High Temp A B A B

PAGE 91

91 Figures 5 14 through 5 16 show various Arrhenius maturity strength relationships developed from the specimens cured under two different curing temperatures for Mix #2. Similarly, for both exposed curing and Burlene covering conditions the use of activat ion energy of 33500 J/mol gives the best matching between the maturity strength relationships generated from the specimens cured under two different curing temperatures. Also, the differences between the calculated equivalent ages at the critical strength of 2200 psi were less than 7% for both exposed curing and Burlene covering conditions. Figure 5 17. Arrhenius maturity streng th relationships with activation energy of 33500 J/mol for Mix #3. A) S pecimens cured under exposed curing condition B) S pe cimens cured with Burlene covering. 0 500 1000 1500 2000 2500 3000 3500 4000 4500 2 5 8 11 14 Compressive Strength (psi) Equivalent Age (hour) Mix #3 Exposed (AE = 33500 J/mol) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 4500 2 5 8 11 14 Compressive Strength (psi) Equivalent Age (hour) Mix #3 Burlene (AE = 33500 J/mol) Med Tamp High Temp Low Temp A B

PAGE 92

92 Figure 5 18. Arrhenius maturity streng th relationships with activation energy of 4 0 000 J/mol for Mix # 3 A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering. Figure 5 19. Arrhenius maturity streng th relationships with activation energy of 45000 J/mol for Mix #3. A) S pecimens cured under exposed curing condition B) S pecimens cured with Burlene covering. 0 500 1000 1500 2000 2500 3000 3500 4000 2 5 8 11 14 Compressive Strength (psi) Equivalent Age (hour) Mix #3 Exposed (AE = 40000 J/mol) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 2 5 8 11 14 Compressive Strength (psi) Equivalent Age (hour) Mix #3 Burlene (AE = 40000 J/mol) Med Tamp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 4500 2 5 8 11 14 Compressive Strength (psi) Equivalent Age (hour) Mix #3 Exposed (AE = 45000 J/mol) Med Temp High Temp Low Temp 0 500 1000 1500 2000 2500 3000 3500 4000 4500 2 5 8 11 14 Compressive Strength (psi) Equivalent Age (hour) Mix #3 Burlene (AE = 45000 J/mol) Med Tamp High Temp Low Temp A B A B

PAGE 93

93 Figures 5 17 through 5 19 show various Arrhenius maturity strength relationships developed from the specimens cured under three different c uring temperatures for Mix #3. Similarly, f or both exposed curing and Burlene covering conditions the use of activation energy of 33500 J/mol gives the best matching between the matu rity strength relationships generated from the specimens cured under three different curing temperatures. Also, the differences between the calculated equivalent ages at the critical strength of 2200 psi were less than 3 % for both exposed curing and Burlen e covering conditions As shown in both parametric studies for both Nurse Saul and Arrhenius maturity method, the maturity strength relationships developed by Arrhenius maturity function with activation energy of 33500 J/mol gave the best strength predicti on under different curing temperatures and curing conditions with error s of less than 7%, 6% and 3% for mix #1, mix #2, and mix #3 respectively. Therefore the Arrhenius maturity function with activation energy of 3350 0 J/mol was use d for the calculation o f maturity ind ices for the further experiment s and field studies 5. 4 Evaluation of Functions for Modeling Maturity Strength Relationship In 1991, Carino conducted a study of comparison s between the three types ( hyperbolic, logarithmic and exponential ) of functions for modeling the maturity strength relationship of concrete He concluded that in the case of his strength and maturity data, the logarithmic function d id not fit the maturity strength data well while t he other two functions show ed a good fit to the maturity strength data Similar comparisons for the three different modeling functions were made in this study to evaluate their suitability for modeling the maturity strength relationship of concre te by using the data from the test result of Mix # 1. Regression analys e s were

PAGE 94

94 performed to relate the compressive strengths to the corresponding equivalent ages of the concrete using these three functions. Since the strength equivalent age relationship works well only at early ages of the concrete only t he strength data at equivalent ages of less than 20 hours were used in these regression analys e s The regression analys e s were performed using MATLAB program. The best fit line for each function was determined by the program by minimizing the sum of the s quare of errors (SSE). The results of these regression analys e s are shown in Table 5 5. Table 5 5 Results of regression analys e s relating c o mpressive strengths to equivalent age of M ix # 1 using three different modeling function s Hyperbolic M odel Expone ntial M odel Logarithmic M odel f(x) = a*(b*(x c)/(1+b*(x c))) f(x) = a*exp( (b/x)^c) f(x) = a+b*log(x) a = 1.067e+004 (8013, 13220) a = 5987 (4953, 7021) a = 2100 ( 2493, 1706) b = 0.04104 (0.02623, 0.05586) b = 9.619 (8.229, 11.01) b = 1855 (1662, 204 9) c = 3.506 (3.251, 3.762) c = 1.335 (1.098, 1.573) SSE: 2.111e+006 SSE: 1.759e+006 SSE: 7.355e+006 R square: 0.9726 R square: 0.9772 R square: 0.9045 Adjusted R square: 0.9717 Adjusted R square: 0.9764 Adjusted R square: 0.9029 RMSE: 186 RMSE: 169. 8 RMSE: 344.4 As can be seen from Table 5 5 the exponential and hyperbolic models gave better fit to the data from mix # 1 with R square values of 0.97 72 and 0.97 26 respectively. Th e logarithmic model gave a poorer fit with the smallest adjusted R squar e value of 0. 9045 and the highest root mean squire error (RMSE) value of 344.4.

PAGE 95

95 Figure 5 20. Results of regression analys e s performed using MATLAB program A) C omparison s of maturity strength curves generated by three different modeling functions f or M ix # 1 B) C omparisons of residuals of the generated maturity strength curves A B

PAGE 96

96 Figure 5 20 shows the comparison of these three functions in modeling the strength equivalent age relationship of entire mix #1 data It can be seen that the exponential and the hyperbolic models give better fit to the data than the logarithmic 5. 5 Evaluation of Curing Environments 5. 5 .1 Comparison of Maturity Strength Plots of Mix # 1 Figure 5 21 shows the c omparison of the maturity strength plots for Mix #1 cured without Burlene covering at three different curing temperatures and cured unde r standard condition. Figure 5 21 Strength vers us equivalent age plots for Mix # 1 cured without Burlene covering a t three different temperatures and standard condition The strength equivalent age plots for the concrete cured under the standard condition are shown by the solid circular black dots on the figure. A best fit hyperbolic line for the strength equivalent ag e relationship for the standard condition is shown as a solid red line, along with two other thinner solid lines showing the upper and lower 0 500 1000 1500 2000 2500 3000 3500 2 4 6 8 10 12 14 16 18 Compressive Strength (psi) Equivalent Age (hour) Exp-Mid Exp-Hot Exp-Cold Standard Std Fit Upper Lower

PAGE 97

97 bounds for 330 psi (15% of 2200 psi) of this prediction line. It can be seen that out of 32 maturity strength po ints plotted, 5 points (or 15.6%) fall slightly above the upper bound and no points fall below the lower bound of this prediction line. Figure 5 22 shows the comparison of the maturity strength plots for Mix #1 cured with Burlene covering at three differen t curing temperatures and cured under standard condition. Similarly, the strength equivalent age plots for the concrete cured under standard condition are shown by the solid circ ular black dots on the figure. A best fit hyperbolic line for the strength equ ivalent age relationship for the standard condition is shown as a solid line, along with two other thinner solid lines showing the upper and lower bounds for 330 psi (15% of 2200 psi) of this prediction line. It can be seen that out of 32 maturity stren gth points plotted, 3 points (or 9.4%) fall slightly above the upper bound and no points fall below the lower bound of this prediction line. Figure 5 22 S trength vers us equivalent age plots for Mix # 1 cured with Burlene covering at three different te mperatures and standard condition 0 500 1000 1500 2000 2500 3000 3500 2 4 6 8 10 12 14 16 18 Compressive Strength (psi) Equivalent Age (hour) Bur-Mid Bur-Hot Bur-Cold Standard Std Fit Upper Lower

PAGE 98

98 Among the 8 points which fall above the upper bound, 6 points were from two separ ate batches (Replicate 1 of Mix # 1 cured in hot curing chamber, and Replicate 2 of Mix #1 cured in ambient condition) that had lower slump a nd air content than batches for standard curing condition. The slump of these two batches of concrete was 6.25 inches, as compared with 6.5 to 6.75 inches and the air content of these two batches of concrete was 2.0%, as compared with 2.2 to 2.3% for t he s tandard batches. The lower slump of the fresh concrete might be due to a lower water content, which would result in a lower water/cement ratio and thus resulting in a higher strength of the concrete. Also, as it has already been shown by previous research results for every 1% increase in total volume of air, a 5% decrease in compressive strength can be expected (Mindess, Young, and Darwin, 2003) L ower air content of fresh concrete result s in hi gher strength of the concrete. For the other batches of concre te which had similar fresh concrete properties, the different curing conditions did not have any significant effects on the strength equivalent age plots at an early age. 5. 5 .2 Comparison of Maturity Strength Plots of Mix # 2 As mentioned previously, some of the concrete batches of Mix # 2 had air content exce eding the specified limit of 6% due to high mixture temperature and water to cement ratio (Wilson et al., 2011). Figure 5 23 shows the comparison of the maturity strength plots for batches containing a n appropriate amount of air to batches containing inappropriate amount of air exceeding the specified limit. Concrete batches for hot and ambient curing temperatures contained appropriate amount of air while batches for standard and low curing temperatures contained inappropriate amount of air.

PAGE 99

99 Figure 5 23 S trength versus equivalent age plots for Mix # 2 containing appropriate and inappropriate amount of air As shown in Figure 5 23, significant strength loss occurred on the batches containing inapprop riate amount of air. The data points for the four batches of concrete with exceedingly high air content were thus excluded and the remaining maturity strength plots are shown in Figure 5 24 Since the data from the standard curing condition had been remov ed (due to high air content ), the data from the concrete cured with Burlene covering and at ambient condition was selected as the data for standard curing condition. The plots for the solid circ ular bla ck dots on the figure. A best fit hyperbolic line for the strength along with two other thinner solid lines showing the upper and lower bounds for 330 psi (15% o f 2200 psi) of this prediction line. It can be seen that all of the 19 maturity strength points lie between the upper and lower bounds. 0 500 1000 1500 2000 2500 3000 3500 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 Compressive Strength (psi) Equivalent Age (hour) Batches with Appropriate Air Content Datches with Inappropriate Air Content

PAGE 100

100 Figure 5 24 S trength versu s equivalent age plots for Mix # 2 cured without Burlene covering at t wo different tempera tures and modified standard condition Figure 5 25. Strength versus equivalent age plots for Mix #2 cured with Burlene covering at two different temperatures and modified standard condition 0 500 1000 1500 2000 2500 3000 3500 2 4 6 8 10 12 14 16 18 Compressive Strength (psi) Equivalent Age (hour) Exp-Mid Exp-Hot Bur-Mid (Standard) Std Fit Upper Lower 0 500 1000 1500 2000 2500 3000 3500 2 4 6 8 10 12 14 16 18 Compressive Strength (psi) Equivalent Age (hour) Bur-Hot Bur-Mid (Standard) Std Fit Upper Lower

PAGE 101

101 For the batches of Mix # 2 cured with Burlene covering, tw o of these batches had air content exceeding 6% and t hese two batches were those cured at 43 F. Similarly, the data from these two batches of concrete were excluded from Figure 5 25 which shows the co mparison of the maturity strength plots for Mix #2 cure d with Burlene covering at two different curing temperatures. Similarly, since the data from the standard curing condition had been excluded (due to high air content), the data from the concrete cured with Burlene covering and at ambient condition were se lected as the data for the standard curing condition. The plots the figure. A best fit hyperbolic line for the strength equivalent age relationship for the lines showing the upper and lower bounds for 330 psi (15% of 2200 psi) of this prediction line. It can be seen that all of the 12 maturity strength points lie between the upper and lower bounds. 5. 5 3 Comparison of Maturity Strength Plots of Mix # 3 Figure 5 26 shows the comparison of the maturity strength plots for Mix #3 cured without Burlene covering at three different curing temperatures and cured under the standard cond ition. The strength equivalent age plots for the concrete cured under the standard condition are shown by the solid circular black dots on the figure. A best fit hyperbolic line for the strength equivalent age relationship for the standard condition is sho wn as a solid line, along with two other thinner solid lines showing the upper and lower bounds for 330 psi (15% of 2200 psi) of this prediction line. It can be seen that all of the 33 maturity strength points lie between the upper and lower bounds.

PAGE 102

102 Figure 5 26 S trength versus equivalent age plots for Mix #3 cured without Burlene covering at three different temperatures and standard condition Figure 5 27 shows the comparison of the maturity strength plots for Mix #3 cured with Burlene covering at t hree different curing temperatures and that cured under the standard condition. Similarly, the strength equivalent age plots for the concrete cured unde r standard curing condition are shown by the solid cir cular black dots on the figure. A best fit hyperb olic line for the strength equivalent age relationship for the standard condition is shown as a solid line, along with two other thinner solid red lines showing the upper and lower bounds for 330 psi (15% of 2200 psi) of this prediction line. It can be seen that all of the 34 maturity strength points lie between the upper and lower bounds. 0 500 1000 1500 2000 2500 3000 3500 2 4 6 8 10 12 14 Compressive Strength (psi) Equivalent Age (hour) Exp-Mid Exp-Hot Exp-Cold Standard Std Fit Upper Lower

PAGE 103

103 Figure 5 27 S trength versus equivalent age plots for Mix #3 cured with Burlene covering at three different temperatures and standard condition 5. 6 Evaluation of Effect of Fresh Concrete Properties The previous section shows that while the Arrhenius maturity method was effective in predicting the strength of the concrete at early age, the variation in fresh concrete properties such as air content and slump could ca use errors in str ength prediction. This section presents the effects of variation of fresh concrete properties on the maturity strength characteristics of the concrete according to the data from the second set of exp eriment s of the laboratory study. 5. 6 .1 Compressive Strength P rediction The effects of fresh concrete properties on the maturity strength characteristics are evaluated by comparing the predicted strength of the concrete at a typical equivalent age for each mix design The equivalent ages, 8, 9 a nd 8 hours, were chosen when the compressive strength approximately reached 2000 psi for Mix # 1, Mix 0 500 1000 1500 2000 2500 3000 3500 2 4 6 8 10 12 14 Compressive Strength(psi) Equivalent Age (hour) Bur-Mid Bur-Hot Bur-Cold Standard Std Fit Upper Lower

PAGE 104

104 # 2 and Mix # 3 respectively. Compressive strengths at the chosen equivalent age s were calculated by interpolation of separately developed hyperbolic trend line s for each curing condition. Table 5 6 5 7 and 5 8 show the developed hyperbolic trend lines and predicted compressive strengths for Mix # 1, Mix # 2 and Mix # 3 respectively Table 5 6. H yperbolic trend lines and predicted strengths calculated by M ATLAB program for Mix # 1 Hyperbolic Trend Lin e f(x) = a*(b*(x c)/(1+b*(x c)) Curing Condition s a b c Sum of Squired Error R S quire d Value RMSE Predicted Strength (psi) Amb 1 Exp 5754 0.1242 4.558 9.61E+04 0.9948 219.2 1723 Amb 1 Cur 6917 0.08405 4 .459 1.03E+04 0.9996 71.67 1586 Amb 2 Exp 6463 0.1185 4.244 3222 0.9999 40.14 1991 Amb 2 Cur. 7807 0.09462 4.399 1.62E+04 0.9995 90.09 1984 113 F 1 Exp 6972 0.1156 4.291 9.03E+04 0.9957 173.5 2092 113 F 1 Cur 8533 0.0738 3.7 3.59E+04 0.9989 109.4 2056 113 F 2 Exp 5993 0.1415 5.081 7.36E+04 0.9964 156.6 175 2 113 F 2 Cur 7639 0.07908 4.431 1.56E+04 0.9995 72.17 1681 43 F 1 Exp 6391 0.08752 3.38 1.95E+04 0.999 98.62 1840 43 F 1 Cur 6989 0.0702 3.425 4.07E+04 0.9982 142.7 169 9 43 F 2 Exp 6204 0.09855 3.107 5.08E+04 0.9972 130.1 2018 43 F 2 Cur 7293 0.06927 3.358 3.76E+04 0.9983 111.9 1774 Standard 1 7366 0.07586 3.962 2.70E+04 0.999 0 116.1 1727 Standard 2 8385 0.06085 4.017 5236 0.9999 51.17 163 6 73 F 1 Exp 6558 0.09453 4.12 3.73E+04 0.9 983 136.6 17 60 73 F 1 Cur 7979 0.06419 3.721 4334 0.9999 46.55 1719 73 F 2 Exp 6838 0.09708 4.195 4.63E+04 0.9982 152.1 184 5 73 F 2 Cur 8666 0.05308 3.949 1545 1 27.8 153 4

PAGE 105

105 Table 5 7. H yperbolic trend lines and predicted strengths calculated by MATL AB program for Mix # 2 Hyperbolic Trend Lin e f(x) = a*(b*(x c)/(1+b*(x c)) Curing Condition a b c Sum of Squired Error R Squired Value RMSE Predicted Strength (psi) Amb. 1 Exp. 6041 0.08419 3.607 1.69E+04 0.9989 91.94 1886 Amb. 1 Cur. 6735 0.08211 3.89 6 3.53E+03 0.9998 42 198 9 Amb. 2 Exp. 6494 0.08297 3.291 3539 0.9998 42.06 2087 Amb. 2 Cur. 7506 0.06668 3.088 2.08E+04 0.9991 101.9 2122 5215 0.1519 3.827 2.60E+04 0.9975 93.07 229 5 7392 0.05411 2.132 3.14E+04 0.9987 102.3 200 3 5288 0.1141 3.145 1.53E+04 0.9986 71.42 211 8 7220 0.05322 1.252 1.73E+05 0.9921 240.3 210 8 5465 0 .0626 3.68 2.74E+04 0.9972 95.55 1365 6127 0.04254 3.111 1.79E+04 0.9983 77.23 1227 5351 0.05403 3.248 2.28E+04 0.9976 87.2 126 9 5962 0.0397 3.002 3.60E+04 0.9965 109.5 114 7 Standard 1 6282 0.04488 2.301 2.96E+04 0.9 982 121.7 1452 Standard 2 6022 0.04807 2.242 4.48E+04 0.9971 149.6 147 7 5955 0.06198 3.032 8687 0.9994 65.9 160 8 6197 0.05749 3.215 7389 0.9995 60.78 154 7 5630 0.05879 2.635 3.33E+04 0.9976 129 1533 5733 0.06062 3.026 4159 0.9997 45.6 1524

PAGE 106

106 Table 5 8 H yperbolic trend lines and predicted strengths calculated by MATLAB program for Mix # 3 Hyperbolic Trend Lin e f(x) = a*(b*(x c)/(1+b*(x c)) Curing Condition a b c Sum of Squired Error R Squired Value R MSE Predicted Strength (psi) Amb. 1 Exp. 6570 0.1376 4.699 2.48E+05 0.9903 352.2 1828 Amb. 1 Cur. 7954 0.0959 4.705 2.93E+05 0.9921 382.5 1910 Amb. 2 Exp. 6816 0.1223 4.632 2.05E+05 0.9925 319.9 1989 Amb. 2 Cur. 8077 0.09477 4.789 3.02E+05 0.9919 388.8 1884 6902 0.1371 4.189 4.68E+04 0.9984 124.9 2369 8300 0.09944 4.188 2.71E+05 0.9938 300.3 2281 6957 0.1351 4.462 8.80E+04 0.9969 171.2 2250 8380 0.09301 4.335 2.35E+05 0.9944 280.1 2130 7729 0.1144 4.148 2.52E+05 0.9923 290 2364 8561 0.08446 4.238 2.81E+05 0.9926 306.2 2064 8271 0.09431 4.153 3.23E+05 0.9911 328.1 2202 8780 0.07597 4.288 3.14E+05 0.9918 323.7 1931 Standard 1 7454 0.07808 3.811 1.40 E+04 0.9995 83.55 1837 Standard 2 7780 0.08632 4.146 5982 0.9998 54.69 1942 6502 0.1112 3.931 6.22E+04 0.9974 176.4 2026 7775 0.07419 3.761 2.22E+04 0.9993 105.2 1860 6957 0.1068 4.211 2.47E+04 0.9991 111.2 2004 F 2 Cur. 7735 0.09853 4.378 7.99E+04 0.9975 199.9 2034 5. 6 2 Effect of Variation of Slump on Maturity Strength Plots Figure 5 28 shows plots of predicted compressive strength at an equivalent age of 8 hours versus the slump of the fresh concrete for Mix # 1. It can be observed that the

PAGE 107

107 compressive strength generally dec reases with increasing slump. As the slump increases by 2 inches, the compressive strength can decrease by as much as 300 psi Figure 5 29 shows plots of compressive strength at an equivalen t age of 9 hours versus the slump of the fresh concrete for Mix #2. Since Mix # 2 had four batches of concrete with exceedingly high air content (as noted in Section 5.2 ), the effect of slump might be over shadowed by the effects of air content, and thus wa s not clearly observed. However, the linear trend line for data f r o m valid batches shows clear trend that follows the general trend as well. Figure 5 28 Plots of compressive strength at equivalent age of 8 hours versus slump of fresh concrete for Mix # 1 y = 109.3x + 2585.9 R = 0.3772 1200 1400 1600 1800 2000 2200 2400 5.5 6 6.5 7 7.5 8 8.5 9 9.5 Compressive Strength (psi) Slump (in)

PAGE 108

108 Figure 5 29 Plots of compressive strength at equivalent age of 9 hours versus slump of fresh concrete for Mix # 2 Figure 5 30 Plots of compressive strength at equivalent age of 8 hours versus slump of fresh concrete for Mix # 3 y = 28.122x + 1837.2 R = 0.005 y = 69.448x + 2405.9 R = 0.5095 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 3.5 4 4.5 5 5.5 6 6.5 7 7.5 Compressive Strength (psi) Slump (in) Valid Mixtures Invalid Mixtures Linear (All) Linear (Valid Mixtures) y = 283.56x + 4262.8 R = 0.6815 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 6.5 7 7.5 8 8.5 Compressive Strength (psi) Slump (in)

PAGE 109

109 Figure 5 30 s hows plots of predicted compressive strength at an equivalent age of 8 hours versus the slump of the fresh concrete for Mix # 3 Similarly, it can be seen that the compressive strength generally decreases with increasing slump. As the slump increases by 2 i nches, the compressive strength can decrease by as much as 500 psi. 5. 6 .3 Effect of Variation of Air Content on Maturity Strength Plots Figure 5 31 shows plots of compressive strength at an equivalent age of 8 hours versus the air content o f the fresh conc rete for Mix # 1 It can be observed that the compressive strength generally decreases as the air content increases for all mixes. The effect of air content of fresh concrete on compressive strength of concrete has been reported by other researchers. It has been reported that for every 1% increase in total volume of air, a 5% decrease in compressive strength can be expected (Mindess et al. 2003). Figure 5 31 Plots of compressive strength at equivalent age of 8 hours versus air content of fresh concrete for Mix # 1 y = 478.24x + 2809.5 1200 1400 1600 1800 2000 2200 2400 1.8 1.9 2 2.1 2.2 2.3 2.4 Compressive Strength (psi) Air Content (%)

PAGE 110

110 Figure 5 32 shows plots of compressive strength at an equivalent age of 9 hours versus the air content of the fresh concrete for Mix # 2 Similarly, i t can be observed that the compressive strength generally decreases as the air content increas es for all mixes. The trend of plots for valid batches also follows a similar general trend as well. Figure 5 32 Plots of compressive strength at equivalent age of 9 hours versus air content of fresh concrete for Mix # 2 Figure 5 33 shows plots of com pressive strength at an equivalent age of 8 hours versus the air content of the fresh concrete for Mix # 3. Similarly, it can be observed that the compressive strength generally decreases as the air content increases for all mixes. y = 231.09x + 3118.4 R = 0.8351 y = 135.92x + 2697.9 R = 0.2714 800 1000 1200 1400 1600 1800 2000 2200 2400 3 4 5 6 7 8 9 Compressive Strength (psi) Air Content (%) Invalid Mixtures Valid Mixtures Linear (All) Linear (Valid Mixtures)

PAGE 111

111 Figure 5 33 Plots of compressive strength at equivalent age of 8 hours versus air content of fresh concrete for Mix # 3 5. 6 4 Effect of Variation of Unit Weight on Maturity Strength Plots Figure 5 34 shows plots of compressive strength at an equivalent age of 8 hours versus the unit weight of the fresh concrete for Mix # 1. It can be observed that the compressive strength generally decreases as the unit weight of the concrete decreases. A decrease in unit weight of concrete can be caused by an increase in air content or water content, which will generally reduce the strength of the concrete. Figure 5 35 shows plots of compressive strength at an equivalent age of 9 hours versus the unit weight of the fresh concrete for Mix #2 Similarly, it can be observed that the compressive strength generally decreases as the unit weight decreases for all mixes. The trend of plots for valid batches also follows a similar general trend as well. y = 532.29x + 3345.6 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2 2.2 2.4 2.6 2.8 3 Compressive Strength (psi) Air Content (%)

PAGE 112

112 Figure 5 34 Plots of compressive strength at equivalent age of 8 hours versus unit weight of f resh concrete for Mix # 1 Figure 5 35 Plots of compressive strength at equivalent age of 9 hours versus unit weight of fresh concrete for Mix # 2 y = 95.152x 12196 R = 0.5301 1200 1400 1600 1800 2000 2200 2400 145 146 147 148 149 150 Compressive Strength (psi) Unit Weight (pcf) y = 136.22x 17355 R = 0.678 y = 49.916x 5012.8 R = 0.4335 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 137 138 139 140 141 142 143 144 145 Compressive Strength (psi) Unit Weight (pcf) Valid Mixtures Invalid Mixtures Linear (All) Linear (Valid Mixtures)

PAGE 113

113 Figure 5 36 shows plots of compressive strength at an equivalent age of 8 hours versus the air content of the fresh concrete for Mix # 3. Similarly, it can be observed that the compressive strength generally decreases as the unit weight decreases for all mixes. Figure 5 36 Plots of compressive strength at equivalent age of 8 hours versus unit weight of f resh concrete for Mix # 3 5. 7 Suggested Testing Protocol s for Generating Maturity Strength Curves Based on the test results, observations, and experience ga ined from the laboratory stud ies the following testing protocol for the use of maturity method to e stimate compressive strength of concrete at early age for slab replacement application is recommended. y = 105.52x 13340 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 143 144 145 146 147 148 Compressive Strength (psi) Unit Weight (pcf)

PAGE 114

114 5. 7 .1 Preparation of C oncrete in the L aboratory The laboratory concrete mixtures which are to be used to develop maturity strength curve s have to be pro duce d using the same mix design and the same time frame to resemble the con crete to be used in an actual project Table 5 9. Recommended time frame for the laboratory concrete Activity Time (minute) Simulating A ctivities for Ready Mix Concrete Mix wi thout Accelerator 5 Concrete m ix in batch plant Slow or Intermittent Mix 15 30 (depends on the distance f ro m batch plant to project site) Delivery time from batch plant to the project site Mix with Accelerator 5 Adding a ccelerator in the project sit e Producing S pecimens Less than 10 Pouring and finishing time of actual slab Table 5 9 shows the recommended time frame determined by an actual slab replacement project The delivery time may vary according to the location of the batch plant and the pr oject site Thus it is recommended to accurately estimate the delivery time before produc ing the concrete mixture in the laboratory. The concrete can also be sampled from the concrete used in the slab replacement project site if the same concrete mix is g oing to be used in the project to be monitored. 5. 7 2 Testing of Fresh C oncrete In order to check the similarity of the laboratory produced concrete to the concrete used in an actual project, slump and air content need to be measured and recorded. The fre sh concrete properties of the laboratory produced concrete have to be tested before adding accelerator to simulate the testing of the concrete used in the actual project. When the fresh concrete properties for laboratory produced concrete are in a specifie d tolerance range of the concrete used to the actual project the concrete

PAGE 115

115 can then be used to generate maturity strength curve s that can be used to predict early age strength of the in place concrete. 5. 7 .3 Instrumenting C oncrete S pecimens Two of the con crete specimens are to be instrumented with temperature loggers to record temperature at a freq uency of once every 10 minutes. To prevent the temperature loggers to be moved during pouring and vibrating concrete they should be fixed at the middle of the c ylinder mold with low temperature conductivity wire and thus accurate temperature recording can be achieved 5. 7 4 Curing of C oncrete S pecimens All the concrete specimens are to be c ured in cylindrical molds. The cylindrical molds are to be covered during the entire time and stored together in the same location. Standard cu ring condition is not needed. The concrete specimens are to be taken out of the molds just right before they are tested for compressive strength. This is done to reduce damage to the conc rete specimens at early age. 5. 7 5 Testing of C oncrete S pecimens A minimum of twelve cylindrical specimens are tested for compressive at four different testing times around the estimated time when the concrete is expected to have a com pressive streng th of 2200 psi. These f our curing times can be 3, 5, 7 and 9 hours, but can be adjusted depending on the type of concrete mix and the curing temperature used. Three replicate specimens are to be tested per curing time. 5. 7 .6 Development of Maturity Streng th R elationship The temperature history of the concrete specimens as recorded by the temperature loggers in the two concrete specimens are to be downloaded to a computer and used to compute the equivalent ages of the concrete at the various times when the

PAGE 116

116 concrete specimens are tested. Since the concrete temperature of the high early age concrete during the curing time should not be dropped below 73 F, an activation energy of 33500 J/mol can be applied for the calculation of equivalent age The compressive strength versus equivalent age data for the tested concrete are to be plotted to develop a relationship between compressive strength and equival ent a ge of the concrete. Both modified hyperbolic and exponential modeling functions can be used for develop ing maturity strength curve. From the developed maturity strength curve the equivalent age of the concrete when compressive strength will be 2200 p si can be determined.

PAGE 117

117 CHAPTER 6 FIELD STUDY 6.1 Testing Plan for Field Study 6.1.1 Overview The main purpose of the field study is to evaluate the effectiveness and reliability of the proposed maturity method for prediction of concrete strength at early age for slab replacement application. Therefore t he proposed testing protocol was applied to multiple actual slab replacement projects in Florida. The field developed maturity strength curve w as compared to th e laboratory developed maturity strength curve for validation of the strength prediction. In addition, recorded temperature histories at the differe nt locations of the actual concrete slab were used to calculate the equivalent age and to predict strengths of the in place concrete at different location s in the slab Figure 6 1 shows the exact locations and dates of the field studies performed in Jacksonville area Figure 6 1 Location s and date s of the field stud ies performed

PAGE 118

118 Table 6 1 shows the mix design of the concrete used for the replaced conc rete slab for the field studies and laboratory produced concrete Table 6 1. Mix design used to both first and second field studies Name Product Name Quantity /Yd of Concrete Specification Cement Type I/II Cement 850 LB AASHITO Type I/II Coarse Aggre gate #57 Stone 1600 LB Fine Aggregate Silica Sand 1101 LB Air Ent Admixture Darex AEA 7.5 OZ AASHITO M 154 AEA Type D Admixture WRDA 60 17.0 OZ AASHITO M 194 Type D Type F Admixture Adva 120 29.8 OZ AASHITO M 194 Type F Type E A dmixture Dara ccel 4 4 8 .0 OZ AASHITO M 194 Type E Water Water 31.2 GA 6.1. 2 Te mperature Measurement s from the I nstrumented Slab s For each field study, 4 temperature loggers were installed after preparation of the base material and before concrete was poured. Figure 6 2 shows the installation s of temperature loggers at the center, corner, longitudinal edge, and transverse edge of the slab and Figure 6 3 shows a close up picture of the installed temperature loggers at corner and transvers e edge of the slab For accura te temperature measurement, plastic spikes that h ave relatively low heat conductivity were used to fix the temperature loggers. The temperature data from the temperature loggers were downloaded on the following day and used to predict compressive strength with the previously developed maturity strength curve. Comparison was made between the predicted strength of the actual slab at the four different locations and the measured strength of the field cured specimens on the replaced slab namely protection specimens in order to evaluate the reliability of protection specimens which are assumed to present the

PAGE 119

119 strength of the actual slab. The temperature loggers were programmed to record the concrete temperature every 10 minutes Figure 6 2 Installed t emperature loggers at four different locations (Photos courtesy of author, Ohhoon Kwon). Figure 6 3 Installed temperature logger s at the corner and edge of the slab (Photos courtesy of author, Ohhoon Kwon).

PAGE 120

120 6. 2 The First Field Study 6. 2 .1 Overview The first field study was performed on Sep 17 2013 at US 1 Alt S Exit to Liberty St ., Jacksonville as shown in Figure 6 1 Two different batches of the ready mix concrete were used at the project site The ready mix concrete used for the 5th concrete sla b , was used to produce specimens at the project site and cure d in the insulated curing box as shown in Figure 6 4 The temperature history of the specimens was recorded and the compressive strength was tested at multiple curing times to develop field generated maturity strength curve. Also, the temperature histories of the 5th slab at four different locations were recorded to evaluate the strength differences at different locations of the concrete slab. Figure 6 5 shows a d rawing for the exact locations of the four temperature loggers The other ready protection was used to produce protection specimen s which was used for evaluating the strength of the in place concrete. The pr otection specimens were cured on the replaced concrete slab with curing blank et (Burlene) covering. The temperature history of the protection specimen s w as also recorded to calculate their equivalent age to develop maturity strength relationship s using the actual strengths of the protection concrete specimens tested by subcontractors.

PAGE 121

121 Figure 6 4 Curing of the specimens produced from the 5 th concrete (Photos courtesy of author, Ohhoon Kwon). Figure 6 5 Drawing for the exact locations of the insta lled temperature loggers The laboratory produced concrete oratory was used to produce specimens in the l aboratory. These laboratory specimens were cured under

PAGE 122

122 ambient laboratory temperature without stripping the cylinder mold s as re commended in the proposed testing protocol. The temperature history of the specimens was recorded and the compressive strength was determined at multiple curing times to generate the laboratory generated maturity strength The generated maturity strength c urves and maturity strength plots were compared to each other to validate the accuracy of the maturity strength prediction under di fferent curing conditions. In a ddition, the predicted strength at four different locations of the 5th conc rete slab were comp ared to the measured strength of the protection specimens. 6.2.2 Development of Maturity Strength Curves In order to validate the strength prediction of the maturity method with application of the proposed testing protocol, multiple maturity strength rela tionships were generated under different curing conditio ns and compared to each other. The time to start placing concrete in the project site and producing specimens in the laboratory was proposed to be a starting point of the curing tim e because temperatu re history could be recorded from this point. However, the prio r elapsed time is also very important to predict accurate concrete strength with maturity method. Thus, all elapsed times for preparation activities of the concrete were recorded to calibrate t he generated maturity strength curve. 6.2.2.1 Laboratory g enerated m aturity s trength c urve To develop a maturity strength relationship in the laboratory maturity strength twelve 4"8" cylindrical specimens for testing compressive str ength and two 4"8" cylindrical specimens for recording temperature history were produced with the same concrete mix design used in the field studies (Table 6 1 ). All specimens were cured in the 73 to 80 F ambient curing room without stripping the cylinder molds.

PAGE 123

123 Compressive strength tests for the twelve specimens were performed at the curing times of 2.83, 5.05, 6.87, and 8.70 hours. T he temperature logge rs in two temperature recording specimens were programmed to record the concrete temperature every 10 minutes and Arrhenius maturity function was used with an activation energy of 33500 J/mol to calcu late equivalent age because concrete temperatur es had not dropped under 73 F during the curing time. 6.2.2.2 Field g enerated m aturity s trength c urve To develop a field generate maturity strength curve for the first field study t en 4"8" cylindrical specimens for testing compressive strength and two 4"8" cylindrical specimens for recording temperature hi story were produced from the 5 th concrete at the project site. Figure 6 6. Compressive strength test for field generated maturity strength curve (Photos courtesy of author, Ohhoon Kwon).

PAGE 124

124 As shown in Figure 6 4 all the field produced specimens were cured in the insulated curing box. Compressive strength of these specimens were measured at five different curing times of 1.50, 1.95, 2.88, 4.07, and 6.00 hours by using a portable compressive strength tester as shown in Figu re 6 6. Similarly, temperature loggers in the two temperature specimens were programmed to record the concrete temperature every 10 minutes and the Arrhenius maturity function with an activation energy of 33500 J/mol was applied to calculate the equivalent ages at the various curing times. 6.2.2.3 Maturity s trength p lots of p rotection s pecimens For the first field stud y, an additional maturity strength relationship was made by recording the temperature history of the protection specimens. Two 4"8" cylindr ical specimens for recording temperature history were made from the protection concrete and placed near the protection spe cimens made by the subcontractors. T he protect ion specimens were cured on newly placed c oncrete slab with curing blanket (Burlene) cov ering. Since the subcontractors measured strength of the protection specimens at only two different curing times, namely 3.78 and 3.91 hours, the matur ity strength curve could not be generated but two maturity strength points were determined Similarly, t emperature loggers in two temperature monitoring specimens were programmed to record the concrete temperature every 10 minutes and Arrhenius maturity function with an activation energy of 33500 J/mol was applied to calculate the equivalent age. 6 2.3 Valid ation of Maturity Strength Prediction Three different maturity strength relationships ( 5 th slab maturity strength curve laboratory maturity strength curve and maturity strengt h plots of protection specimens) were compared to validate the accuracy of matu rity strength prediction. As explained previously, all maturity strength relationships were developed under totall y different

PAGE 125

125 curing conditions. Thus, if the three maturity strength relationships are similar to one another, it can verify the postulation th at different curing conditions do not have any effect on the maturity strength prediction and the maturity method would give a reliable early age strength prediction for concrete used in slab replacement project s Two 4"8" cylindrical specimens were used to record temperature histories for each batch of concrete and the average temperature value s from the temperature histories were used to calculate equivalent age s Figure 6 7 shows the recorded temperature histories for the three batches of concrete used in the first field study Figure 6 7 R ecorded temperature histories for the specimens produced by different batches of concrete Table 6 2 shows the strength test results and calculated equivalent ages for the specimens produced from three different b atches of concrete and curing conditions. The specimens produced from the 5th concrete were cured in the insulated curing tank without temperature control and the specimens produced from the protection concrete 40 60 80 100 120 140 160 180 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 Elapsed Time (hour) Specimen 1_5th Concrete Specimen 2_5th Concrete Protection Specimen 1 Protection Specimen 2 Specimen 1_Lab Concrete Specimen 2_Lab Concrete

PAGE 126

126 were cured on the replaced slab with curing blanket covering. Lastly, the concrete specimens produced from the lab oratory concrete were cured under ambient laboratory temperature without stripping cylinder mol ds. Since the curing conditions were different, the concrete specimens produced by the sam e mix design had totally different strength and equivalent ages at the similar elapsed curing time. T he time of placement was applied to calculate the equivalent ages as a starting point of the curing time. Table 6 2. Compressive strength test results and corresponding equivalent ages for the t hree batches of concrete Specimens P roduced by the 5th C oncrete Elapsed C uring T ime ( h our) Compressive Strength (psi) Equivalent Age (hour) Curing Condition 1.50 249 2.97 Cured in the water filled curing tank wi thout temperature control 1.95 687 4.20 2.88 1934 6.58 4.07 2420 8.66 6.00 3526 12.13 11.53 6203 22.11 Specimens P roduced by the P rotection C oncrete Elapsed C uring T ime ( h our) Compressive Strength (psi) Equivalent Age (hour) Curing Condition 3 .78 2210 8.12 Cured on the replaced slab with curing blanket covering 3.91 2300 8.34 Specimens P roduced by the L aboratory C oncrete Elapsed C uring T ime ( h our) Compressive Strength (psi) Equivalent Age (hour) Curing Condition 2.83 2 95 3.7 7 C ured under a mbient laboratory temperature without stripping of their cylinder molds 5.05 2168 8.07 6.87 32 24 11. 59 8.71 3854 14. 46

PAGE 127

127 Table 6 3 shows the best fit regression equations for the generated maturity strength relationships. Since the specimens produced from the protection concrete do not have enough strength data for generating maturity strength curve, only two maturity strength curves for the other two batches of concrete were generated. MATLAB program was used to calculate the best fit maturity streng th curves with a modified exponential function which was recommen ded by ASTM C 1074 The maturity strength curves were determined by the program by minimizing the sum of the square of errors (SSE). As can be seen from Table 6 3 both maturity strength curv es had a good correlation between the actual strength and the calculated equivalent age with R square values of higher than 0.99. Tab le 6 3. Results of regression analys e s for two maturity strength relationships with modified exponential function 5 th S l ab M aturity S trength Curve L aboratory M aturity S trength Curve f(x) = a*exp( (b/x)^c) f(x) = a*exp( (b/x)^c) a = 7 046 a = 6097 b = 8. 848 b = 8. 436 c = 1. 097 c = 1. 449 SSE: 40920 SSE: 102.2 R square: 0. 99 42 R square: 1 Adjusted R square: 0. 9 883 Adjust ed R square: 1 RMSE: 143 RMSE: 10.11 Figure 6 8 shows the comparison s of developed maturity strength curves and plots for three different batches of concrete. It can be seen that the maturity strength curve generated by the 5th concrete has approximat ely 7 % higher strength at the critical strength range of 2000 to 2500 psi than the other two batches of concrete. On

PAGE 128

128 the other hand, the other two batches of concrete, namely the protection concrete and lab oratory concrete, have identical maturity strength plots at the critical stren gth range of 2000 to 2500 psi. Figure 6 8 Developed maturity strength curves from different batches of concrete before waiting time adjustment Table 6 4 shows the recorded times for the preparation of concrete batches at the project site and laboratory. As shown in Table 6 4, the 5th concrete had waited for more than 20 minutes after adding accelerator because of the sudden heavy rain at the project site. This longer waiting time allowed more time for hydration of cemen t before its temperature history was recorded and it possibly can be explained for the different strength prediction of the maturity strength curve generated by the 5th concrete On the other hand, the same time frame was applied for the preparation of t he protection concrete and lab oratory concrete to valida te the accuracy of maturity strength prediction for the protection specimens. As a result, strength predictions for the protection 0 500 1000 1500 2000 2500 3000 3500 4000 0 3 6 9 12 15 Compressive Strength (psi) Equivalent Age (hour) Field_5th Concrete_Strength Field_5th Concrete_Model Laboratory_Strength Laboratory_Model Field_Protection_Strength

PAGE 129

129 specimens made by the laboratory generated maturity strength curve we re very accurate with prediction errors of 3. 0 % and 2.6%. Table 6 4 Recorded time for preparation of batches of concrete Activity 5 th Concrete (Sep 17 2013 ) Protection Concrete (Sep 17 2013 ) Activity Laboratory Concrete (Sep 2 6 2013 ) Time Elapse d Time (minute) Time Elapsed Time (minute) Time Elapsed Time (minute) Mix Start 10 :46 pm 0 11:45 pm 0 Mix Start 1 :11 p m 0 Depart f ro m Batch Plant 11:00 pm 14 11:57 pm 12 Slow Mix Start 1:16 am 5 Arrive at Field 11:15 pm 29 00:08 am 23 Add Accele rator 1:41 am 30 Add Accelerator 11:20 pm 34 00:1 3 am 2 8 Producing Specimens 1:50 am 3 9 Concrete Placement 11:43 pm 57 00:23 am 38 6.2.4 Comparisons of the S trength Prediction s at Different L ocation s of Concrete Slab cal specimens cured under identical curing condition of the in place concrete slab are used to estimate strength of the concrete slab in the slab replacement project. However, it is well known that different shape and volume of concrete structures have dif ferent rate of strength development and thus, it is hard to generalize that strength of the protection specimens accurately repre sent the actual strength of the in place concrete. In this section the strength of the in place concrete at different location s and protection specimens were compared to determine the strength differences between different locations of in place concrete and protecti on specimens. Since different batches of concrete were used for the 5th slab and protection specimens, appropriate c uring time adjustments for the 5 th concrete was applied.

PAGE 130

130 Figure 6 9 shows comparisons of th e temperature time plots for different locations of concrete slab and protection specimens. It can be see n that different locations of the concrete slab and protect ion specimens have totally different trends of temperature histo ries and it would result in different rates of strength development. Figure 6 9 C omparisons of the temperature time plots for different locations of concrete slab and protection specimen s Figure 6 1 0 shows comparisons of the predic ted strength time plots for different locations of the 5th slab and protection specimens. S trength predictions for both the 5th slab and protection specimens were made by previously generated maturity strength curve from lab oratory concrete. Similarly, the curing time adjustment s were made on the predictio ns of strength for different locations of the 5th slab. As can be observed in Figure 6 1 0 the strength of the protection specim ens was slightly lower than the lowest strength of t he in place concrete at early age. However, at later than the curing time of 3.3 hours, the strength of the protection specimens shows slightly higher strength than 40 60 80 100 120 140 160 0 1 2 3 4 5 6 7 8 Elapsed Time (hour) 5th Slab Corner 5th Slab Center 5th Slab Edge Long. 5th Slab Edge Trans. Ave. Prot. Specimens

PAGE 131

131 lowest strength of the in place concrete and it may cause early openin g of the replaced concrete slab to traffic. Figure 6 1 0 C omparisons of the predicted strength time plots for different locations of concrete slab and protection specimens 6. 3 The Second Field Study 6. 3 .1 Overview The second field study was performed on Oct. 9 at US 1 Alt S Exit to Pearl St., Jacksonville as shown in Figure 6 1. As planned a batch of the ready mix concrete used for the last concrete slab last was used to produce specimens at the project site and cured on the last concrete slab with curing blank et (Burlene) covering to resemble the curing condition of the protection specimens The temperature histor y of the specimens was recorded and the compressive strength was determined at multiple curing times to develop field g enerated maturity strength curve for the second field study slab maturity strength Also, the temperature histories of the 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 0 1 2 3 4 5 6 7 8 Compressive Strength (psi) Elapsed Time (hour) 5th Slab Corner 5th Slab Center 5th Slab Edge Long. 5th Slab Edge Trans. Ave. Prot. Specimens

PAGE 132

132 last slab at four different locations were recorded to evaluate the strength differences at different locations of the concrete slab and protection specimens Since the same mix d esign had been used for both the first and second field studies, t he laboratory generated maturity strength curve used in the first field study was applied to the second field study as well. Bot h field and laboratory generated maturity strength curves were compared to each other to validate the accuracy of maturity strength prediction under di fferent curing conditions. In addition, predicted strength at four different locations of the last con cre te slab were compared to strength of the protection specimens. 6. 3 .2 Development of Maturity Strength Curve To develop a field generated maturity strength curve for the second field study, eight 4"8" cylindrical specimens for testing compressive strength and two 4"8" cylindrical specimens for recording temperature history were produced from the last concrete at the project site. All field produced specimens were cured on newly placed last concrete slab with curing blanket (Burlene) covering. Compressive s trengths of these specimens were measured at four different curing times of 2.57, 3.70, 5.20, and 7.48 hours by using a portable compressive strength tester as shown in Figure 6 6. The temperature loggers in two temperature monitoring specimens were progra mmed to record the concrete temperature every 5 minutes and the Arrhenius maturity function with an activation energy of 33500 J/mol was used to calculate the equivalent ages at the various curing times. Similar to the first field study the time to start placing concrete for the last slab was used as a starting point of the curing time.

PAGE 133

133 6 3 .3 Validation of Maturity Strength Prediction The field generated maturity strength for the second field study was compared to the laboratory maturity strength used in the first field study to validate the accuracy of the maturity strength prediction. Since both maturity strength curves were developed under different curing conditions, if the trend s of the maturity strength curves are identical it can verify the postula tion that different curing conditions do not have any effect on the maturity strength prediction at early age and the maturity method could give a reliable early age strength prediction for concrete use d in slab replacement project s Figure 6 1 1 R eco rded temperature histories for the specimens produced by different batches of concrete Figure 6 1 1 shows the recorded temperature h istories for the batches used in the second field study. Two 4"8" cylindrical specimens were used to record the 40 50 60 70 80 90 100 110 120 130 140 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 Elapsed Time (hour) Specimen 1_Last Concrete Specimen 2_Last Concrete Specimen 1_Lab Concrete Specimen 2_Lab Concrete

PAGE 134

134 temperature histories for each batch of concrete and the average value of the temperature histories was used to calculate the corresponding equivalent age. Table 6 5 Compressive strength test results and corresponding equivalent ages for two batches of concrete L ast Concrete Elapsed C uring T ime ( h our) Compressive Strength (psi) Equivalent Age (hour) Curing Condition 2.57 481 3.91 Cured on the replaced slab with curing blanket covering 3.70 1599 6.89 5.20 2963 10.40 7.48 3911 13.41 Laboratory Concrete El apsed C uring T ime ( h our) Compressive Strength (psi) Equivalent Age (hour) Curing Condition 2.83 295 3.77 C ured under ambient laboratory temperature without stripping of their cylinder molds 5.05 2168 8.07 6.87 3224 11.59 8.71 3854 14.46 Table 6 5 shows the strength test results and calculated equivalent ages for the specimens produced from the last concrete and the lab oratory concrete. T he specimens produced from the last concrete were cured on the replaced slab with curing blanket covering and the concr ete specimens produced from lab oratory concrete were cured under ambient laboratory temperature without stripping cylinder mol ds. A s shown in Table 6 5, c oncrete specimens made from two different batches of concrete had different strength and equ ivalent ages at the similar elapsed curing time due to the different curing conditions. Similarly, t he time of placement for both batches of concrete was used as the starting point of curing time in the calculation of the equivalent ages

PAGE 135

135 Tab le 6 6 Resul ts of regression analysis for two maturity strength relationship with modified exponential function Last Slab Maturity S trength Curve Laboratory Maturity S trength Curve f(x) = a*exp( (b/x)^c) f(x) = a*exp( (b/x)^c) a = 1 4020 a = 6097 b = 18.22 b = 8. 4 36 c = 0.793 c = 1.4 49 SSE: 514.4 SSE: 102.2 R square: 0.99 99 R square: 1 Adjusted R square: 0.9 998 Adjusted R square: 1 RMSE: 22.68 RMSE: 10.11 Table 6 6 shows the best fit regression equations for the generated m aturity strength relationships. MAT LAB program was used to calculate the best fit maturity strength curves with a modified exponential function which was recommended by ASTM C 1074. The maturity strength curves were determined by the program by minimizing the sum of the square of errors (S SE). As can be seen from Table 6 6 the regression equations for both maturity strength curves compare well to the plots of actual strength versus calculated equivalent age with R square values of higher than 0.99. Figure 6 1 2 shows the comparison of the d eveloped maturity strength s f ro m the last concrete and the laboratory concrete It can be seen that both batches of concrete have an identical trend at the critical stren gth range of 2000 to 2500 psi.

PAGE 136

136 Figure 6 1 2 Developed maturity strength s from d ifferent batches of concrete before waiting time adjustment Table 6 7 Recorded time for preparation of batches of concrete Activity Last Concrete ( Oct. 9 2013 ) Activity Lab oratory Concrete ( Sep. 26 2013 ) Time Elapsed Time (minute) Time Elapsed T ime (minute) Mix Start 2:02 am 0 Mix Start 1:11 p m 0 Depart f ro m Batch Plant 2:10 a m 8 Slow Mix Start 1:16 p m 5 Arrive at Field 2:25 a m 23 Add Accelerator 1:41 p m 30 Add Accelerator 2:44 am 42 Producing Specimens 1:50 p m 39 Concrete Placement 2:50 a m 48 Table 6 7 shows the recorded times for the preparation of concrete batches at the project site a nd laboratory. Due t o unexpected delay at the project site, the last concrete had waited for almost 10 more minutes than the laboratory concret e. However, s trength prediction made by the last slab maturity strength was identical to strength 0 500 1000 1500 2000 2500 3000 3500 4000 4500 0 2 4 6 8 10 12 14 Compressive Strength (psi) Equivalent Age (hour) Field_Last Concrete_Strength Field_Last Concrete_Model Laboratory_Strength Laboratory_Model

PAGE 137

137 prediction made by laboratory maturity strength because the acce lerator was not added during the additional waiting time before the concrete was placed 6. 3 4 Comparisons of the S trength Prediction s at Different L ocation s of Concrete Slab For the second field study, the temperature histories of the concrete in the last slab we re monitored at four different location s and the temperature histories of the protection specimens were also monitored The predicted strengths of the protection specimens and the in place concrete at four different locations were compared to each other to determine the s trength differences Figure 6 1 3 Curing of the protecti on specimens and the last concrete slab (Photos courtesy of author, Ohhoon Kwon). As shown in Figure 6 1 3 protection specimens were cured on the last concrete slab to estimate strength for the concrete slab by testing the protection specimens Figure 6 1 4 shows comparisons of th e temperature time plots for the in place concrete at different locations of the replacement slab and the protection specimens. T hough the

PAGE 138

138 in place concrete and the protection specimens seemed to cure under the same curing conditi on, the in place concrete at different locations of the concrete slab and the protection specimens have totally different trends of temperature histories These different temperature histories would result in different rates of strength development. Fi gure 6 1 4 Comparisons of the temperature time plots for the concrete at different locations of slab and the protection specimens Figure 6 1 5 shows comparisons of the predicted strength time plots for the concrete at different locations of the last slab and the protection specimens. S trength predictions for both the in place concrete and the protection specimens were made from the maturity str ength curve for the lab oratory concrete. 40 50 60 70 80 90 100 110 120 130 140 0 1 2 3 4 5 6 7 8 Elapsed Time (hour) Last Slab Corner Last Slab Center Last Slab Edge Long. Last Slab Edge Trans. Ave. Prot. Specimens

PAGE 139

139 Figure 6 1 5 Comparisons of the predicted strength time plots for the concrete at different locations of slab and the protection specimens As can be observed in Figure 6 1 5 the strength of the protection specimens is almost two times higher than lowest strength of the in place at early age. At the project site the su bcontractors had to wait until the time when the protection specimens reached a compressive strength of 2200 psi before they opened the concrete slab to traffic in accordance with FDOT slab replacement specification s However at th e time to open traffic, the strength of the in place concrete at the slab corner was less than 1200 psi Thus, using the strength of the protection specimens as strength determination may result in over prediction of concrete strength and result in too early opening of the replac ement slab to traffic. 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 0 1 2 3 4 5 6 7 8 Compressive Strength (psi) Elapsed Time (hour) 5th Slab Corner 5th Slab Center 5th Slab Edge Long. 5th Slab Edge Trans. Ave. Prot. Specimens Time to open to traffic

PAGE 140

140 6.4 Curing Time Adjustment for Field Generated Maturity Strength Curve 6.4.1 Overview To be able to make concrete strength prediction in the field using the maturity method, the maturity strength curve for the concrete must be availa ble prior to the placement of the concrete in the field As observed in both field studies, different time frames for preparation of the laboratory and field concrete could result in error s on maturity strength prediction. Therefore, it is recommended to u se an accurately estimated time frame for use in the preparation of the laboratory concrete with considerations of possible factors including batching and delivery time. When more or less t ime has been taken to place concre te at the project site than the estimated time frame which was applied to produce the laboratory concrete it is also recommended to adjust the curing time on the maturity strength curve generated from the laboratory concrete by adding or subtracting the amount of equivalent age caused by the time difference. 6.4. 2 Proposed Curing Time Adjustment Throughout both field studies total of three maturity strength curves namely the 5 th slab maturity strength the laboratory maturity strength and last slab maturity strength were obtained. T he batches of concrete used to generate the maturity strength curves were produced and cured in totally different environments and thus, the generated maturity strength curves can be used to verify the accuracy of maturity s trength prediction. Table 6 8 s hows the fresh concrete properties right befor e adding accelerator. Since three di fferent batches of concrete had very similar fresh concrete properties, three maturity s trength curves developed by three batches of concrete could be applied for the maturit y strength prediction.

PAGE 141

141 Table 6 8. F resh concrete properties of the three batches of concrete Batches of Concrete 1 st Field Study 5 th Slab (Sep. 17, 2013) 2 nd Field Study Last Slab (Oct. 9 2013) Laboratory Concrete (Sep. 26, 2013) Specification (FDO T Mix N o. 353 128) Slump (in) 3.0 2.5 3.5 1.5 4.5 Air Content (%) 1.9 N/A 2.1 1.0 6.0 Figure 6 1 6 shows the comparison of the three maturity strength curves obtained from both field studies The developed maturity strength curves were compar ed to one another and used to predict strength of the protection specimens tested by subcontractors for the first field study. Figure 6 1 6 Comparisons of the different maturity strength curves and plots of protection specimens M aturity strength pred ictions for the protection specimens made by the laboratory maturity strength and the last slab maturity strength seem to have reliable accuracy with less than 3 % of prediction error s However the 5 th slab maturity strength shows more than 6% of predictio n errors on its strength prediction. 0 500 1000 1500 2000 2500 3000 3500 4000 0 2 4 6 8 10 12 14 Compressive Strength (psi) Equivalent Age (hour) 1st_5th Slab_Strength 1st_5th Slab_Model 2nd_Last Slab_Stregnth 2nd_Last Slab_Model Laboratory_Strength Laboratory_Model Actual Protection

PAGE 142

142 In general, ready mix concrete is placed as soon as the concrete mixer truck arrives at the project site. However, as shown in Figure 6 1 7 the 5 th concrete had waited for more time after adding accelerator than the ot her concrete This longer waiting time would allow more time for hydration of cement before its temperature history was recorded and it could explain for the bigger error on strength prediction made by the 5 th slab maturity strength Figure 6 1 7 Prepa ration time frames for the different batches of concrete On the other hand more waiting time of the last concrete seeme d not to have any effect on maturity strength prediction This can be explained that the hydration rate of the type I/II cement withou t the accelerator is much lower than the hydration rate of the cement with the added accelerator Thus, it can be explained that a negligible amount of hydration took place in the last concrete during additional waiting time ( Set Accelerator 2012). 34 42 30 28 23 6 9 10 0 10 20 30 40 50 60 1st_5th Concrete 2nd_Last Concrete Laboratory Concrete Protection Concrete Elapsed Time (minute) Different Batches of Concrete Elapsed Time to Adding Accelerator Elapsed Time to Placement

PAGE 143

143 To m inimize matur ity prediction error due to unexpected waiting time at the project site it is recommended to incorporate the waiting time elapsed from the time of adding accelerator by adjust ing the curing time s of the concrete in the application of the matu rity strength curve As shown in Figure 6 1 8 three different batches of concrete ( 5 th concrete, last concrete, and laboratory concrete ) had waited for 1 3 4 and 1 more minutes than the waiting time of the protection concrete. T hese time difference s can be converted to the equivalent age s of 0. 39 0. 10 and 0.0 2 hour s, and add ed to the equivalent ages of their maturity strength curves. Figure 6 1 8 Comparisons of developed maturity strength curves and plots obtained from both field studies after ad justment of different waiting times Figure 6 1 8 shows c omparisons of developed maturity strength s after adjustment of different waiting time It can be observed that all maturity strength curves predict strength of the protection specimens with lower tha n 3% of prediction errors. 0 500 1000 1500 2000 2500 3000 3500 4000 4500 0 2 4 6 8 10 12 14 Compressive Strength (psi) Equivalent Age (hour) 1st_5th Slab_Adjusted 2nd_Last Slab_Adjusted Laboratory_Adjusted Actual Protection

PAGE 144

144 Table 6 9 shows comparison s of the accur acy of strength predictions for the protection specimens having strength of 2210 and 2300 psi at the curing time of 3.78 and 3.87 hours respectively made by different maturity strength c urves with and without curing time adjustment Table 6 9. Maturity strength predictions made by different maturity strength curves 5 th Slab Maturity S trength Curve Last Slab Maturity S trength Curve Laboratory Maturity S trength Curve Strength (psi) Pre diction Error (%) Strength (psi) Prediction Error (%) Strength (psi) Prediction Error (%) Without Time Adjustment 23 69 7.20 21 24 3.89 2143 3.03 2 455 6.74 2222 3.39 2 241 2.57 With Time Adjustment 2232 1.00 2 102 4.89 2135 3.39 2321 0.91 2 195 4.57 2234 2.87 It can b e seen that the accuracy of strength predictions made by the 5 th slab maturity strength with curing time adjustment were remarkably improved with prediction errors of 1 % and 0.91% as compared with the strength predictions m ade by the 5 th slab maturity strength without curing time adjustment. On the other hand, no remarkable changes were observed for strength prediction made by the other two maturity strength s. Though the proposed curing time adjustment did not consider the e lapsed time prior to the add ition of accelerator, it made a remarkable reduction of the prediction error for the 5 th slab maturity strength Also all strength predictions made by the adjusted maturity strength s had a very reliable accuracy with prediction errors of less than 5% at the critical strength range of 2000 to 2500 psi.

PAGE 145

145 CHAPTER 7 CONCLUSIONS AND RECOMMENDATIONS 7 .1 Summary of Findings 7.1.1 The F ir st Set of Laboratory E xperiment s The main objective s of the first set of laboratory experiment s w e re to evaluate the effectiveness of four different maturity systems, namely Humboldt, Command Center, Intelli Rock, and COMA meter under various curing temperatures and to select the most appropriate one to be used in the maturity strength prediction. Als o, the specimens specimens were evaluated to determine the mo st effective specimen size to be used for the rest of the study As a result of t he first set of experiment s the Command Center maturity system set of experiment s and field studies. The decision was made based on t he possible factors that can affect the maturity strength prediction such as accuracy, resolution, response time c on venience of the maturity system, trend of temperature history, and strength of the different size s of specimens The main findings from the first set of experiment s in this study are summarized as follows: The Intelli Rock and Command C enter temperature logger s appear ed to show the greatest accuracy with little or no detectable error. The Humboldt temperature thermocouples appear ed to give an error of 2 to 4F. The Command Center temperature logger ha d a resolution of 1F, while the Inte lli Rock and Humboldt sensor s/loggers ha d a resolution of 2F. The Intelli Rock temperature logger s appear ed to have a longer reaction time (delay time) than the Command Center and Humboldt sensors /loggers

PAGE 146

146 The Intelli Rock temperature logger s were relati vely larger in size than the Command Center and Humboldt sensors /loggers and the embedded Intelli Rock temperature logger s ha d a negative effect on strength of the concrete specimen s Other temperature sensors /loggers we re relatively small as compared to the concrete specimens So, strength of the concrete specimens we re not significantly affected by the other embedded temperature sensors /loggers The Intelli Rock temperature logger s need ed to be pre set at the factory for their frequency of temperature reading, while the Command Center logger s c ould be set by the user conveniently. The Intelli Rock and Command Center maturity logger s w ere more convenient to use in the field than the Humboldt meter, since the temperature logger s d id not need to be connect ed to the maturity meter continuously during the test. The COMA meter was very easy and convenient to use. However, it ha d to be read manually and provide d only measurements in equivalent age of concrete using a fixed reference of 20C and a fixed activati on energy of 40 000 J/mol. we re 10 to 12 specimens at the peak point hour earlier than ens for the concrete mixes used in this study The different trends of the temperature histories fo r both size of specimens cause d different trend s of the maturity time plots but there were no differences detected in the generated maturity strength relationship s 7.1.2 The S econd Set of Laboratory E xperiments The main objective s of the second set of ex periment s were to evaluate the possible effects of different placement and curing environments on the predicted strength of concrete from the maturity method and to determine the most appropriate procedure to be used to obtain accurate predicted strength o f concrete. Two maturity functions, namely the Nurse Saul and Arrhenius maturity functions were evaluated. Also the effect s of different curing environment s and fresh concrete properties were evaluated to achieve the goal of the second set of experiment s

PAGE 147

147 As a result the Arrhenius maturity function with an activation energy of 33500 J/mol was chosen to be used for maturity strength prediction because it showed consistent maturity strength relationships developed by various groups of specimens cured unde r various curing conditions. It was also found that different curing conditions d id not have any significant effect on the Arrhenius maturity strength relationships at early age while different fresh concrete properties ha d a remarkable effect on the matur ity strength relationships. Therefore, it is not recommen ded to use maturity strength prediction when the fresh concrete properties of the laboratory generated concrete exceed the specification limit of concrete used in the actual project. Based on the te st results, observations, and experience gained from the second set of experiment s, the following testing protocol for the use of maturity method to estimate compressive strength of concrete at early age for slab replacement application is recommended : The same concrete ingredients mix design and preparation time of concrete as used in the actual field project should be used for the laboratory produced concrete mixture which is used to generate the maturity strength curve The fresh concrete properties such as slump and air content should be measured right before the addi tion of accelerator to simulate the testing of fresh concrete at the project site. The laboratory generate d maturity strength curve can be used to predict concrete strength when the fres h concrete measurement s are in the specified tolerance range. Two of the concrete specimens are to be instrumented with temperature loggers to record temperature at a frequency of once every 10 minutes. The s tandard curing condition is not needed. The con crete specimens are to be taken out of the molds just right before they are tested for compressive strength. A minimum of twelve cylindrical specimens are tested for compressive at four different testing times around the estimated time when the conc rete is expected to have compressive strength of 2200 psi.

PAGE 148

148 The equivalent age for developing maturity strength is recommended to be calculated by the Arrhenius maturity function with an activation energy of 33500 J/mol 7.1.3 Field Study The mai n objectiv e of the field study wa s to evaluate the effectiveness and reliability of the proposed maturity method for prediction of concrete strength at early age for slab replacement application. The field developed maturity strength curves were compared to the labo ratory developed maturity strength curve to validate the accuracy of maturity strength predictio n. Also predic ted strength of the in place concrete at different location s of the replacement slab were compared to the actual strength of the protection speci mens to evalua te the reliability of using the strength of cyli ndrical specimens as the estimated strength of the concrete in the slab. T he results of the study indicate that the maturity strength prediction show ed great accuracy when the same concret e preparation time were applied for both concrete batches, namely the concrete batch used to develop maturity strength curve and the other batch used in the replacement slab When more or less t ime ha s been taken to place concrete at the project site than the estimated preparation time which was applied to the concrete batch used to develop the maturity strength relationship, it is recommended to a djust maturity strength relationship s to be used for the concrete at the project site by adding or subtracting the amount of maturity index caused by the time difference. Also in both field studies the protection specimen s showed higher strength than strength of the concrete at the slab corner at the time to open the slab to traffic Thus, using strength of the protection specimens as strength determination may result in over

PAGE 149

149 prediction of concrete strength and result in too early opening of the replacement slab to traffic. 7 .2 Conclusions The use of maturity concept to predict compressive strength of the concret e slab for slab replacement project at early age was found to be a reliable and convenient strength prediction method Some limitations of maturity strength prediction such as the strength loss due to high curing temperature and insufficient moisture suppl y were observed during the various laboratory studies performed in this study. However, these limitations were observed at the later age when compressive strength reached around 3000 to 3500 psi, and thus the observed limitations d id not have any negative effect on early age strength prediction of the concrete in the replacement slab. T he proposed testing protocols to apply maturity method for estim ating concrete strength was developed as a result of this study A ppropriate adjustment procedure for the fie ld developed maturity strength relationship to minimize maturity s trength prediction error due to unexpected waiting time was also proposed and verified to have worked well from the results of multiple field studies. Based on the test results, observation s, and e xperience gained from the laboratory experiments and field studies the use of maturity method to estimate compressive strength of concrete at early age for slab replacement application is recommended 7 .3 Recommendations Since this study aims to p ropose the most appropriate application procedure for maturity strength prediction for the concrete slab at early age, most of the analys e s and observations of the test results were mainly focus ed on the curing time of 3 to 8 hours

PAGE 150

150 when the high early stre ngth concrete used to the slab replacement reache d a critical strength of 2200 psi. Therefore maturity strength prediction with the proposed procedures is recommended to use for prediction of early age strength of concrete used in slab replacement project s The rate of the strength development of concrete mix es used in the laboratory experiments and field studies were remarkably higher than the rate of strength development of normal concrete at early age Therefore small differences of the preparation ti me for the batches of concrete used to generate maturity strength relationship in the lab oratory and used to the actual slab in the field may cause a substantial error of maturity strength predictions. Thus, the same time frame for the preparation of both batches of concrete is recommended to be use d When the time difference occurs in the field due to uncontrollable reasons, the proposed time adjustment is recommended to be appl ied to the maturity strength relationship.

PAGE 151

151 APPENDIX TEST RESULTS ON THE MI X #2 IN THE FIRST SET OF EXPERIMENT S Figure A 1 Temperature laboratory condition Figure A 2 Variations of temperature measurements from same temperature sensors /loggers cured under ambi ent laboratory condition 70 75 80 85 90 95 100 105 110 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt Room temp 0 1 2 3 4 5 6 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt

PAGE 152

152 Figure A 3 Temperature time plots of 6 12 laboratory condition Figure A 4 Variations of temperature measurements from same temperature sensors /loggers cured under ambient labor atory condition 65 75 85 95 105 115 125 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt Room temp 0 1 2 3 4 5 6 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt

PAGE 153

153 Figure A 5 Temperature time plots of 4 8 F environment control chamber Figure A 6 Variations of temperature measurements from same temperature sensors /loggers cured in 113 F environment control chamber 80 90 100 110 120 130 140 150 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt Room temp 0 1 2 3 4 5 6 7 8 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt

PAGE 154

154 Figure A 7. Temperature curing tank Figure A 8 Variations of temperature measurements from same temperature sensors /loggers cured in standard curing tank 70 75 80 85 90 95 100 105 110 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt Room temp 0 1 2 3 4 5 6 0 5 10 15 20 25 30 Elapsed Time (hour) Intelli Rock A Intelli Rock N Command Center Humboldt

PAGE 155

155 Figure A 9. E quivalent ages from COMA meter and Intelli Rock maturity meter under three different curing conditio ns 0 5 10 15 20 25 0 20 40 60 80 100 120 140 160 180 Equivalent age (day) Elapsed time (hour) COMA Ambient 4 in Intelli rock Ambient 4 in COMA Ambient 6 in Intelli rock Ambient 6 in COMA 113 Chamber 4in Intelli rock Chamber 4 in

PAGE 156

156 LIST OF REFERENCES Cement Ratios on the Compressive Strength and Workability of Concrete and Lateritic Concrete The Pacific Journal of Science and Technology 12(2), 99 105. Alexander, K. M. and Taplin, Australian Journal of Applied Science 13, 277 284. Temperature of Freshly Mixed Hydraulic Conshohocken, PA. Conshohocken, PA. American So West Conshohocken, PA. od for Slump of Hydraulic Conshohocken, PA. Am Conshohocken, PA. Making Conshohocken, PA. Conshohocken, PA. Bagheri Zadeh, S., Kim, H., Hounsell, S., Wood, C. R., Soleymani, H., and King, M. Journal of Construction Engineering and Management 133(11), 827 835. California Department of Transportatio

PAGE 157

157 Carino, N. J. and Handbook on Nondestructive Testing of Concrete, 2 nd Ed., CRC Press: Boca Raton, FL., 101 146. Carino, N. J., Lew, H. S., and Volz, C. K. (1983 ACI Journal 80(2), 93 101. Materials Laboratories Rep. No. C 310 Research and Geolog y Division, Bureau of Reclamation, Denver, CO. rolled Journal of the Nordic Concrete Federation 1, 21 25. Report of the Director of Research Portland Cement Assn, Skokie, IL., 1 49 163. ACI Materials Journal 86(4), 341 353. Magazine of Concrete Research, 1(1), 21 28. Mindless, S., Young, J. F., and Darwin, D. (2003). Concrete 2 nd Ed., Pearson Education, Inc., Upper Saddle River, NJ. Transportation Research Record: Journal of the Tran sportation Research Board 1900, National Research Council, National Academy Press, Washington, D. C., 79 85. t hesis, Auburn University, Aub urn, AL. Research Report for ALDOT, Contact No. 930 590 Auburn University, Auburn, AL. Magazine of Concrete Research 1(2), 79 88.

PAGE 158

158 Magazine of Concrete Research 6(17), 79 92. Atmospheric Pressu Magazine of Concrete Research Mar., 127 140. S et A < http://www.targetproducts.com/UserContent/SpecSheets/setacc.pdf > (Oct. 23, 2013). Soutsos, M. N., Situ Strength Assessment of Concrete Proc., 5 th Int. Conf., NDT in Civil Engineering, T. Uomoto ed. Tokyo: Elsevier Science, 583 592. Struble, L., and Hawkins, P. (1994). Significance of Tests and Properties of Concrete and Concrete Making Materials 4 th Ed., Fredericksburg, VA, 449 461. ACI Material Journal 88(1), 74 83. Transportation Research record 1775, Transportation Research Board, Washington, D. C., 125 132. Tia, M. a nd ion of Early Strength Requirement of Concrete for Slab Replacement Using APT Research Report for FDOT, Contract No. D099558 University of Florida, Gainesville, FL. Van Dam, T. J., Peterson, K. R., Sutter, L. L., Panguluri, A., Sytsma, J., Buc h, N., Kowli, Opening to Traffic Portland NCHRP Report 540 Transportation Research Board, Washington, D.C. Wade, S. S., Schindler S. K., Barnes, R. W., and Ni Research Report for ALDOT, Contact No. 930 590 Auburn University, Auburn, AL. Master thesis, Auburn University, Auburn, AL. "Florida." (2013). < http://en.wikipedia.org/wiki/Florida > (Sep. 25, 2013). Wilson, M. L. and Kosmatka, S. H. (2011). Design and Control of Concrete Mixtures 15 th E d., Portland Cement Assn, Skokie, IL.

PAGE 159

159 BIOGRAPHICAL SKETCH Ohhoon Kwon was born in Daegu, South Korea. He received his B.S. in architectural engineering from Yeung nam University, South Korea in 2004. While seeking his B.S., he served in the Korean milita ry for two years (1999 2001). He earned his M.S. in architectural engineering from Yon sei University, South Korea in 2006. After his graduation, he worked for Lotte Construction and Engineering Company for two years (2006 2008) as a construction manager. In 2010, he started working toward his Ph.D. degree in civil engineering at the University of Florida, Gainesville. Since 2011, he has become a fellow of the International Road Federation (IRF). In spring 2012, during the time of his Ph.D. plan, he was awa rded M.S. in civil engineering from the University of Florida. His academic advisor was Dr. Mang Tia and his research experiences include d E valuation of L ong L ife C oncrete P avement P ractices for U se in Florida (FDOT, DBK75 977 48) and S lab R eplacement M aturity G uidelines (FDOT, DBK75 977 62) He received his Ph.D. from the University of Florida in the fall of 2013