Accelerated Slab Replacement Using Temporary Precast Panels and Self Consolidating Concrete Steven C

Florida State University Libraries

Electronic Theses, Treatises and Dissertations The Graduate School

2018 Accelerated Slab Replacement Using Temporary Precast Panels and Self Consolidating Concrete Steven C. Squillacote Jr.

Follow this and additional works at the DigiNole: FSU's Digital Repository. For more information, please contact [email protected] FLORIDA STATE UNIVERSITY

FAMU-FSU COLLEGE OF ENGINEERING

ACCELERATED SLAB REPLACEMENT USING TEMPORARY PRECAST PANELS AND

SELF CONSOLIDATING CONCRETE

STEVEN C. SQUILLACOTE, JR

A Thesis submitted to the Department of Civil and Environmental Engineering in partial fulﬁllment of the requirements for the degree of Master of Science

2018

Kamal Tawﬁq Professor Directing Thesis

Michelle Rambo-Roddenberry Committee Member

Lisa Spainhour Committee Member

Raphael Kampmann Committee Member

The Graduate School has veriﬁed and approved the above-named committee members, and certiﬁes that the thesis has been approved in accordance with university requirements.

ii I would like to dedicate this to Dr. Kamal Tawﬁq.

iii ACKNOWLEDGMENTS

I would like to thank Dr. Kamal Tawfiq for your continual support. You hired me as an undergrad student for this research five years ago and I couldn’t be more grateful for that. You have not only been my boss, but also my mentor and friend. I would like to thank Dr. Jamshid Armaghani for supporting this research and pushing it to the federal level. I would like to thank the ladies in the front office. Mrs. Morris, thank you for helping with all the paperwork that goes with being a graduate student. Thank you Mrs. Daynah Blake for saving my day and putting a smile on my face. I cannot thank you ladies enough for your help and your motivation to finish this. The Florida Department of Transportation (FDOT) is gratefully acknowledged for providing the financial support for this study. Also thanks to Argos, A-Materials, BASF, for their assistance and material supply. Dr. Kampmann and Dr. Roddenberry, thank you both for motivating me to finally finish this project. And finally, I would like to thank my beautiful girlfriend, Lauren Beckwith. Without her help, I would have never finished this.

iv TABLE OF CONTENTS

List of Tables ...... vii List of Figures ...... viii Abstract ...... x

1 Introduction 1 1.1 Research Purpose ...... 2 1.2 Research Objectives ...... 2 1.3 Research Scope ...... 3 1.4 Report Organization ...... 3

2 Literature Review 4 2.1 Introduction ...... 4 2.2 Precast Concrete Pavements (PCP) ...... 4 2.3 Self-Consolidating Concrete (SCC) For Pavements ...... 7 2.4 Maturity Method ...... 8 2.4.1 Hydration of Concrete ...... 9 2.4.2 Concrete Curing ...... 10 2.4.3 Concrete Maturity ...... 10 2.4.4 Strength – Maturity Relationship ...... 12

3 Methodology 14 3.1 Introduction ...... 14 3.2 Phase One ...... 14 3.2.1 Development of the SCC Mix ...... 14 3.2.2 SCC Mix Ingredients ...... 15 3.2.3 Trial Mixes ...... 17 3.2.4 SCC Testing – Plastic Phase ...... 18 3.2.5 Development of Precast Replacement Slabs ...... 19 3.2.6 Testing of Precast Replacement Slabs ...... 19 3.3 Phase Two ...... 23 3.3.1 Location ...... 23 3.3.2 Preparation of Replacement-Slab Test-Pit ...... 23 3.3.3 Construction of Temporary Precast Panels ...... 27 3.3.4 Installation of Precast Panels ...... 31 3.3.5 Instrumentation and Test Loading the Panels ...... 33 3.3.6 Removal of Precast Panels from the Test Pit ...... 35 3.3.7 SCC Mix - Field Evaluation ...... 35 3.3.8 Veriﬁcation of Workability Retention at FAMU-FSU Laboratory ...... 38 3.3.9 Casting of Replacement Slab in Test Pit at Test Track ...... 39 3.4 Phase Three ...... 40 3.4.1 Location and Test Setup ...... 40

v 3.4.2 Instrumentation ...... 41 3.4.3 Concrete Pouring ...... 42 3.4.4 Concrete Curing ...... 42 3.4.5 Cylinder Testing ...... 42

4 Results and Analysis 44 4.1 Introduction ...... 44 4.2 Phase One Results ...... 44 4.2.1 Fresh Concrete Properties ...... 44 4.2.2 Hardened Concrete Properties ...... 45 4.2.3 Precast Panel Test Results ...... 48 4.3 Phase Two Results ...... 48 4.3.1 Fresh and Hardened Concrete Properties ...... 49 4.3.2 Truck Loading of Replacement Slab ...... 55 4.3.3 Precast Panels Test Results ...... 56 4.3.4 Truck Loading on Precast Panels ...... 58 4.4 Phase Three Results ...... 62 4.4.1 Fresh and Hardened Concrete Properties ...... 62 4.4.2 Development of Maturity Curves ...... 63

5 Discussion and Conclusion 73 5.1 Project Summary ...... 73 5.2 Conclusions ...... 74 5.2.1 Self-Cosolidating Concrete (SCC) Mix ...... 74 5.2.2 Temporary Reusable Precast Panels ...... 75 5.2.3 Nondestructive Strength Testing (Maturity Curves) ...... 76 5.3 Recommendations for Further Study ...... 76 5.3.1 General ...... 76 5.3.2 Self-Consolidating Concrete (SCC) Mix ...... 77 5.3.3 Temporary Reusable Precast Panels ...... 78 5.3.4 Nondestructive Strength Testing (Maturity Curves) ...... 79 5.4 Recommendations for Future Research ...... 79

Bibliography ...... 80 Biographical Sketch ...... 82

vi LIST OF TABLES

2.1 ASTM C150 Types of Concrete ...... 9

2.2 Properties of Cement Compounds, (Sidney Mindess, 2003) ...... 10

2.3 Hydration of Portland Cement (Camp, 2015) ...... 11

3.1 Basic Mix Design, Initial Mix Design Without Any Modiﬁcations ...... 15

3.2 Six Trial Batches With Modiﬁcations ...... 16

3.3 Coarse and Fine Aggregate Properties, #57 Limestone and Silica Sand Produced by Vulcan...... 17

3.4 Admixture Descriptions ...... 18

3.5 Standard ASTM Tests for Fresh Concrete Properties ...... 20

3.6 General Acceptance Criteria for SCC Workability ...... 21

3.7 SCC Final Mix Design ...... 36

3.8 Final SCC Mix Design Argos Versus A-Materials ...... 41

4.1 Ambient Temperatures and Fresh Concrete Test Results ...... 45

4.2 Compressive Strength Test Results (psi) ...... 47

4.3 Compressive Strengths and Weights of Vibrated vs. Non-Vibrated Samples of TM1 . 47

4.4 Slump Flow and T-20 Test Results ...... 50

4.5 Test Results at Argos Plant Trial Batches ...... 52

4.6 Fresh Concrete Properties for Argos Mix ...... 62

4.7 Hardened Concrete Properties Argos Practice Mix 1 ...... 63

4.8 Hardened Concrete Properties Argos Mix 2 and Mix 3 ...... 64

4.9 Example Maturity Data from A-Materials ...... 65

4.10 Final SCC Mix Design Argos Versus A-Materials ...... 69

4.11 Maturity Predictions for Argos Mix 4 and A-Materials Mix 5 ...... 70

4.12 Novel Approach Maturity Comparison ...... 71

vii LIST OF FIGURES

2.1 Precast Slab Design For Test Pavement Section, (Larson and Hang 1972) ...... 6

2.2 Akashi Kaikyo Bridge Anchorage (Okamura, Ouchi)...... 9

3.1 Precast Panel with Test Pit ...... 22

3.2 Field Test on the Precast Panels ...... 22

3.3 Pavement Test Track ...... 23

3.4 Forms For the Replacement-Slab Test-Pit ...... 24

3.5 Paving Concrete Around the Replacement Slabs Test Pit ...... 25

3.6 Test Pit With Fine Recycled Concrete Aggregate as Leveling Course ...... 25

3.7 Temporary Precast Panels for 6’ x 12’, 8’ x 12’, 10’x 12’ and 12’x 12’ Replacement Pits 26

3.8 Precast Slab Design ...... 27

3.9 Lift Anchors in the Precast Panels ...... 28

3.10 Locations of Covered Lift Anchors ...... 29

3.11 Casting of Precast Panel ...... 29

3.12 Striped Panel Bottom and Backer Glued Baker Rod Along the Panel Sides ...... 30

3.13 Lubricated Backer Rod ...... 30

3.14 Rigging and Maneuvering the Precast Panel ...... 31

3.15 Alternative Rigging Methods ...... 32

3.16 Use of a Four-Sling-Lift Method and a Boom Crane at FAMU-FSU Test Site . . . . . 32

3.17 Placement of the Two Precast Panels ...... 33

3.18 Instrumentation of the Precast Panels at the Test Track ...... 34

3.19 Front and Rear View of the Truck Used in Load Testing of Precast Panels ...... 35

3.20 Damage to One of the Precast Panels During Removal ...... 37

3.21 Temperature Measurement of SCC Mix ...... 39

3.22 Temperature Reading Devices (Tawﬁq and Armaghani, 2016) ...... 41

viii 4.1 Temperature Measurement of SCC Mix ...... 46

4.2 Compressive Strength of Vibrated and Non-Vibrated Cylinder Samples ...... 47

4.3 Vertical Displacement of Precast Panels ...... 48

4.4 Horizontal Displacement of Precast Panels ...... 49

4.5 Trial Batches at Argos Concrete Plant in Jacksonville, Fl ...... 51

4.6 Addition of the Accelerator at the Track ...... 53

4.7 Surface Finish of Replacement Slab ...... 53

4.8 Slab Curing with Plastic Sheet ...... 54

4.9 Load Testing of the Replacement Slab Using a 25,000 lb Truck ...... 55

4.10 Core Samples From the Replacement Slab – No Sign of Aggregate Segregation . . . . 56

4.11 Schematic Plan of Instrumentation of the Precast Panels ...... 57

4.12 Vertical and Horizontal LVDTs on the Precast Panels ...... 57

4.13 Concrete Pump Truck Loading Precast Panels ...... 58

4.14 Conﬁguration of LVDTs and Wheel Loads ...... 59

4.15 Vertical Displacements of the Two Precast Panels ...... 60

4.16 Horizontal Displacement of Precast Panel 2 ...... 60

4.17 Proﬁle of Vertical Displacement of Precast Panel 1 ...... 61

4.18 Imprints of Ridged Strips on Surface of Recycled Aggregate Leveling Layer ...... 61

4.19 Argos 28 Day Maturity Curve ...... 65

4.20 Argos Maturity Curve (First 12 Hours) ...... 66

4.21 28 Day A-Materials Maturity Curve ...... 67

4.22 12 Hour A-Materials Maturity Curve ...... 67

4.23 Argos Theoretical Maturity Values ...... 68

4.24 A-Materials Theoretical Maturity Values ...... 69

4.25 Maturity Relationship from Diﬀerent Sensors Mix 1 ...... 71

4.26 Maturity Relationship from Diﬀerent Sensors Mix 4 ...... 72

ix ABSTRACT

As it stands, many of Florida’s roads have already reached their designed service life and are now in the process of being renewed. The current method in rehabilitation of concrete pavement requires the expired piece of pavement to be cut and removed, place new dowel bars, and then epoxied into the surrounding slabs. Once the slab area has been prepared, fresh concrete is poured, and finished. The concrete is then cured and monitored to achieve a strength requirement of 2,200 psi in the shortest possible time before the lanes can be opened for traffic. This event has been known to take a long time and on major highways lane where lane closure may not exceed 8 hours. This restriction limits the number of slabs that can be replaced. The types of concrete used on these projects are also problematic. In the past, high amounts of cementitious material was used and this can lead to premature cracking. To improve production levels, accelerate construction time at a reduced cost, and provide long lasting pavement, the current research study presents an alternative method of using precast slab panels and self-consolidating concrete. This was accomplished by testing several SCC mixes in the laboratory to achieve concrete with high workability without, high early strength and without segregation. Then, precast panels were designed and built for quick installation and removal. This study also necessitated full scaled field tests where precast slab panels with the proper SCC mix were used. The slabs were tested by a loaded truck moving over it repeatedly and the slab was monitored for any movement and displacements caused by driving and braking on it. After the data was collected from the precast panels, the slabs were then removed and fresh SCC was then poured into the empty pit. The SSC slab was left to cure and the maturity of the concrete was monitored to achieve the required strength for lane opining. In this study, three techniques were used to monitor the concrete maturity. These techniques involved the use of the conventional thermocouples, thermal camera, and laser gun. The traffic load was then applied by driving a dump truck loaded to 25000 pounds over the track for 100 laps. The SCC mix behaved as designed and presented in this study. It achieved a high workability and retained a high slump for nearly an hour. It also exceeded the required FDOT strength requirement of 2200 psi for lane opening. The precast panels proved to be highly durable during the installation, testing, removal and can be reused for other similar applications. Results from

x this study proved proved that using this method has several beneﬁts including greater productivity, reduced maintenance of traﬃc, shorter project completion time.Further, it may reduce the case of premature cracking due to the increase amount of curing time.

xi CHAPTER 1

INTRODUCTION

One of Florida’s top issues, in regard of the transportation industry is that many concrete pavement roads have exceeded their designed service life. The current method of restoring the expired pavement is to either re-mill the surface to give the pavement a renewed smooth like finish, or the other option is to replace the retired segment of pavement. Slab replacement can be done on a small portion of an existing slab or the entire segment can be replaced. The current method of doing so includes many steps such as, removing the deteriorated segment/full slab, repairing the base, drilling holes in the surrounding concrete to inset dowel bars, and then pouring fresh concrete into the pit. The concrete must have high strength early in the curing process so that the lane can be opened back to traffic. The current required compressive strength of concrete needed before the traffic can be reopened is set by Florida Department of Transportation (FDOT) at 2,200 psi in their Specification 353. The biggest problem with the current Slab replacement method is that it’s very time consuming. This is a problem because the time a lane can be closed is restricted to when traffic is at its lightest which is typically at night. This gives about 8 hours of work time to place barricades, cut up the slab, haul it away, repair the base, install the dowel bars and pour fresh concrete. The concrete must be poured early enough to allow proper curing, which is generally 4 hours before the lanes reopen. This process does not allow for many slabs to be replaced in one night, as a result, the timeline to replace many slabs is long. This leads to a high overall cost. Another issue is that the concrete being used is prone to early cracking; this is mainly due to the high amounts of cement and accelerator in the concrete mix. Many of the mixes used for slab replacement exceed 1,000 pounds of cement per cubic yard which typically increases the shrinkage rate of the concrete leading to the formation of cracks.

1 1.1 Research Purpose

The FDOT wanted to improve the current practice of slab replacement. FDOT supported this project to develop a method to use temporary precast panels along with self-consolidating concrete to accelerate the current slab replacement process. This would accomplish higher productivity rate along with a higher quality concrete that is less susceptible to shrinkage cracking. They also wanted to reduce the amount of time required for the maintenance of traffic during construction and reduce the overall project cost. The accelerated slab replacement method involves using precast slabs that are produced and stored off site. When ready these panels are hauled by truck to the project site. Unlike the traditional slab replacement method that takes place in a single night, the accelerated slab replacement method could spread the work over several nights. To do this the areas that have deteriorated concrete roadway a crew can come in as they would before and remove the existing concrete and repair base. However, they do not prep the area for new concrete instead they bring in the precast slab to fill in where the deteriorated concrete panel was originally. With the precast panel set in place, the traffic can be reopened. This allows for more than one area to be prepped for concrete. When all retired slabs have been removed and replaced with precast panels the panels can be quickly removed to allow a crew to finish prepping the pits. The pits are prepped by drilling holes and fastening dowel bars with epoxy. After the pits are ready fresh concrete can be poured. In this method self-consolidating concrete (SCC) would be used to fill the pits to form a permanent replacement slab.

1.2 Research Objectives

The objective of this research was to develop a method to accelerate the traditional method of replacing dilapidated concrete roadways using a two-component replacement system that allows for a higher amount of productivity and a reduced project cost. This project also called for the development of Self-Consolidating Concrete (SCC) mix that mitigates the chance of shrinkage cracking and can cure fast enough to meet FDOT’s Specification 353.This specification states that the compressive strength must meet 2200 psi before the lane can be opened to traffic. Another objective was to create method of how to make this type of concrete using different sources for the materials in a real world setting.

2 1.3 Research Scope

The scope of this project was divided into three interrelated segments along with a preceding literature review to ensure a proper approach throughout all stages of this venture. The first stage was to develop a SCC mix and test it in according to ASTM specifications. Data was gathered for both fresh and hardened properties. This included unit weight, spread, temperature, mix time, and hourly compressive strength. There was also testing of precast replacement slabs. This slab was tested to see how it deflects as loads are applied. It was used to test different bumper materials to help protect itself from sustaining damage when it was lowered into a pit to simulate the actual process of installing precast panels. The second stage is very similar to the first stage just on a larger scale. This stage was in combination of lab testing and field work. This phase tested how to properly install the precast slabs. The SCC mix was also tested on a larger scale. It was made off site at a local batching plant and delivered of a prepared pit to simulate how it would be used in the commercial industry. Finally, the last stage developed a method to predict the SCC mixes strength without the use of destructive concrete testing. A nondestructive testing method was ideal in conjunction with self- consolidating concrete. This was due to the short amount of curing time required before lanes were opened to traffic. During the research, it was difficult to transport concrete samples to a location that had the ability to test samples at designated times. To overcome this issue, the method chosen to determine concrete strength was the Maturity Curve method.

1.4 Report Organization

Following the research scope, Chapter 1 serves as an introductory of the research and it is followed by Chapter 2 which discusses the literature review. Chapter 3 outlines the methodology of how this research was completed as described in the research scope. Thereafter, Chapter 4 discusses results and analysis; the conclusion of the project is presented in Chapter 5.

3 CHAPTER 2

LITERATURE REVIEW

2.1 Introduction

The use of precast concrete technology and SCC mixes in pavement construction and rehabilitation is relatively new when compared with their use in structural applications. The need for rapid solutions to concrete paving construction and replacement of deteriorated slabs has focused attention on precast technology and SCC. Besides accelerated process of repair, precast construction offer many benefits such as long life expectancy. This is because they are casted and cured in a controlled environment that allowed a low water-to-cementitious-materials ratio and a high level of uniformity, which minimizes the potential for premature cracking due to traffic loads. The SCC mix also has several benefits such as high workability, that allows rapid discharge and casting of replacement slabs, and after casting the concrete it is designed to meet high early strength requirements for lane opening to traffic. The development of a method to incorporate both SCC and precast panels along with establishing a Maturity curve to accelerate slab replacement will help contractors with productivity, accelerated completion time, and possible cost savings.

2.2 Precast Concrete Pavements (PCP)

Precast concrete pavements (PCP) have been used in rehabilitation projects as permanent replacements or overlays for long continuous sections of concrete pavements, or in isolated individual or group slabs. The PCP technology includes precast post-tensioned slabs for continuous sections (Merritt et al. 2003), or precast reinforced panels for applications in isolated individual or consecutive slabs (Buch et al. 2003) and (Fort Miller, www.fortmiller.com). The use of PCP started back in the 1930s, however it was not until recently that several U.S highway agencies have begun to implement it. In the early years, the technology was explored either as a matter of curiosity or research (Tayabji). According to Raymond Rollings and Yu T.Chou, who published an article in 1981, the ﬁrst use of precast slabs was around 1931-1932; the Soviet Union constructed unreinforced concrete hexagons that had 4.1 feet long sides and ranged in thickness

4 from 3.9 to 5.5 inches. It was also mentioned that these hexagons tended to rock and spall which is a type of concrete deterioration, and around 1950, when more modern concrete placing equipment became available, the hexagons were replaced with reinforced rectangular concrete slabs (Rollings and Chou). In 1968, the South Dakota Department of Highways and the Federal Highway Administration built a 24 ft-wide, 900 ft-long test section of precast, prestressed concrete slabs on U. S. Highway 14 near Brookings, South Dakota (Larson and Hang 1972). The pavement design was based on South Dakota State University research sponsored by the South Dakota State Department of Highways and Federal Highway Administration (Gorsuch 1962, Kruse 1966, Jacoby 1967, and Hargett 1970). The final slab design used in construction, shown below in Figure 2.1. was 6 ft wide, 24 ft long and 4-1/2 in. thick. The slabs were prestressed using 3/8 in. diameter longitudinal cables and were prestressed to provide 400 psi prestress in the slab. The concrete slabs were overlaid with asphaltic concrete, the depth of which varied from 3-1/2 in. at the road center to 1-1/2 in. at the edge to provide the required surface slope and smoothness. The slabs were swung into placed by a crane in which the area the slabs were being installed on a thin layer of sand around a half inch thick. Once in place, the slabs were seated by a vibratory roller (Rollings and Chou). The layout of the slabs consisted of half of the slabs placed with the long side parallel to the direction of the traffic. The remaining slabs were placed perpendicular to the direction of traffic without connection joints. This South Dakota project was used as an example to recommend precast construction for airport pavements (Hargett 1969). In 1981, 116 damaged concrete panels were replaced at Lindberg Airfield in San Diego California (Rollings and Chou). The damaged slabs were removed and replaced with precast reinforced concrete slabs. Before the new slabs could be installed the old panels were removed and six inches of subgrade was excavated, concrete was then poured until the desired height of the bottom of the precast replacement slabs had been reached. The precast slabs were lowered into placed and seated by a 10 ton roller. This method allowed the airport to stay in operation because there was no time need to allow the concrete to cure. The pavement was strengthened by 8 inch thick asphaltic concrete overlay after all repairs were completed. (Rollings and Chou). Today there are guidelines for precast panels. For example, in 2009 a specification was put into affect for tollways in Illinois that covered the requirements for precast concrete pavement slab

5 Figure 2.1: Precast Slab Design For Test Pavement Section, (Larson and Hang 1972)

6 systems in regards to material and fabrication. This guideline had designers submit several items to the Tollway for review in which included drawings, installation instructions, material, and hardware used. After the review, the system designer is required to perform a trial installation. After the trail, the slabs can be manufactured. The manufactured slabs have to be made by a certiﬁed precast producer and inspected in accordance to the Illinois Department of Transportation’s Manual for Fabrication of Precast Prestressed Concrete Products (Illinois department of transportation).

2.3 Self-Consolidating Concrete (SCC) For Pavements

Self-consolidating concrete (SCC) has properties that differ considerably from conventional concrete. SCC mixes have very high workability and a flow rate that makes it easy to flow through densely reinforced and complex structures, filling in all void without segregation (Yang). Self- consolidating concrete is used primarily in cast-in-place or precast structural members with highly congested reinforcement. Because of this, conventional SCC mixes are usually designed using small size aggregates (grade 89 or smaller). Small size aggregates increases the passing ability of the SCC and the typical aggregate size to opening size relationships are at least at a ratio of 1:2 (Yang). For an SCC mix to achieve a high flow rate that is segregation resistant and can retain very high workability for an extended period of time, admixtures such as high range water reducers, set retarders and workability retainers are used. In fact, there might be a delay in the mix setting time due to the high dosage rates of high range water reducing admixtures, retarders and viscosity modifiers. The use of SCC started in Japan in the early 1980s. At this time it was identified that the lack of uniform and complete compaction of concrete was responsible for poor performance of concrete structures. In an attempt to remove the problem of compaction, research was started at the University of Tokyo by Professor Hajime Okamura, whom developed the first practical SCC mix. By 1990s the innovated concrete mix had spread to Europe. In recent years, SCC mixes are becoming more widely used in construction due to its favorable attributes, such as productivity improvements, reduced labor costs, improved work environment and safety, and improved final product quality. The cast-in-place applications of SCC are usually limited to applications such bridges, buildings, drilled shafts and tunnel linings. However, there

7 have not been applications of SCC in highway pavements and limited research has been conducted to study such use. One such research had been conducted at the Iowa State University, National Concrete Pave- ment Technology Center. The goal was to develop a new type of SCC for slip-form paving to simplify construction and make smoother pavements (Wang 2011). The project was split into two phases; the first phase was the feasibility phase in which the researchers developed an SCC mix that would be adapted to slip-form paving. The mix had to possess sufficient strength while still in a plastic state, also known as “green” Strength (Wang 2011). In this “green strength” state the concrete would be shaped as it passed through the slip forms. In the second phase, additional research was completed on the effects of different materials and admixtures on rheology (Wang 2011). The results indicated that SCC in a semi flowable state can be used in slip form technology without losing its shape after extrusion. This was done on a bike path in Ames, Iowa and after 3 years of service the path showed no cracks. Another research article by Hajime Okamura and Masahiro Ouchi indicated that SCC can be used in large scale construction. The Akashi Straits Bridge was used as an example which opened in 1998 and was the suspension bridge with the longest span in the world. SCC was used in the construction of the two anchorages, shown in Figure 2.2, of the bridge. The concrete was mixed next to the site and was pumped 200 meters to the casting site. It was noted that no segregation occurred and that using SCC shorten the construction time by 20% while using 50 workers instead of 150 (Okamura, Ouchi). Research by Khayat (Khayat and Mitchel 2009) developed guidelines for the use of SCC in precast, prestressed concrete bridge elements. These guidelines address the selection of constituent materials, proportioning of concrete mixtures, testing methods, fresh and hardened concrete properties, production and quality control issues, and other aspects of SCC. Many of these recommendations can be adopted for cast-in-place applications of SCC.

2.4 Maturity Method

To fully understand the concept of concrete maturity, a brief history must be reviewed. The history of the Kansas Test Method KT-44 has roots that are linked to two common maturity functions (Nurse-Saul and Arrhenius). Knowledge of hydration process and the curing of concrete

8 Figure 2.2: Akashi Kaikyo Bridge Anchorage (Okamura, Ouchi). will give further light on how the maturity works. All these variables rely on each other and can aﬀect the strength of concrete

Table 2.1: ASTM C150 Types of Concrete

Cement Type Description Type I Normal Type II Moderate Sulfate Resistance Type II (MH) Moderate Heat of Hydration Type III High Early Strength Type IV1 Low Heat Hydration Type V High Sulfate Resistance

2.4.1 Hydration of Concrete

The chemical reaction that occurs when cement is hydrated is an exothermic process; this is a result of tricalcium silicate in the cement coming into contact with water. Several different types of cement exist and are classified in ASTM C150-99 as shown in Table 2.1. Besides Tricalcium aluminate, there are a couple of other chemical compounds that help cement cure. Table 2.2 lists other commonly found chemicals that are found in Portland cement. Each of these chemicals react at different stages of the hydration of cement. The hydration process can be broken down into

9 Table 2.2: Properties of Cement Compounds, (Sidney Mindess, 2003)

Chemical Compound Description

Tricalcium aluminate, C3A It contributes a lot of heat during the early stages of hydration, but has little strength contribution.

Tricalcium silicate, C3S This compound hydrates and hardens rapidly. It is largely responsible for Portland cement’s initial set and early strength gain.

Dicalcium silicate, C2S This compound hydrates and hardens slowly. It is largely responsible for strength gain after one week

Ferrite, C4AF It hydrates rapidly, but does not contribute much to strength of the cement paste

five different stages (Camp, 2015). Each stage is described in Table 2.3. The first stage of cement hydration is the rapid heat generation of the reaction between water, calcium, and hydroxide ions.

2.4.2 Concrete Curing

For most concretes the highest rate of heat generation occurs in stages I through IV. The Na- tional Ready Mix Concrete Association (NRMCA, 2000) states that curing is maintaining adequate moisture content and temperature in concrete in the earlier stages so that it can develop the wanted properties. Proper curing of concrete is essential to alleviate cracks, increase strength and decrease permeability (PCA, 2015). Evaporation of mix water must also be kept to a minimal by either using a water curing method such as sprinkling water or sealing the concrete samples with plastic. Water curing must be done as soon as possible to avoid the sealing of capillary pores due to drying. Conversely, concrete cannot be water cured too early; this can lead to a rise in the water cement ratio and possibly damage the concrete.

2.4.3 Concrete Maturity

Maturity started back in the 1950’s when it was believed that a correlation of time and temperature could summarize the curing stages of concrete. After Nurse (1949) ﬁrst pioneered the concept of maturity another researcher Saul (1951) reviewed the work and suggested that maturity should be calculated with respect to a datum temperature (To), (Datum temperature is the lowest state

10 Table 2.3: Hydration of Portland Cement (Camp, 2015)

Stage Title Description 1 Rapid heat generation Upon mixing with water, calcium and hydroxide ions are released from the surface of the C3S; pH rises to a very alkaline solution. When the calcium and hydroxide reach critical concentrations, crys- tallization of CH and C-S-H begins. Early chemical reac- tions are temperature dependent, (15 minutes).

2 Dormant Period Causes cement to remain plastic (2-4 hours). The reaction slows. CH crystallizes from the solution; C-S-H develops on the surface of the C3S and forms a coating. As the thickness increases, the time it takes water to penetrate the coating increases, thus the rate of reaction becomes dif- fusion controlled. C2S at a slower rate because it is a less reactive compound.

3 Acceleration Period Critical concentration of ions is reached and silicate hydrates rapidly, maximum rate occurs at this stage. Final set has passed and early hardening begins (4-8 hours).

4 Deceleration Rate of reaction slows; completely diﬀusion dependent reaction.

5 Steady State Constant rate of reaction (12- 24 hours). Temperature has little eﬀect on hydration at this point.

11 in which concrete cannot gain strength). This resulted in the Nurse-Saul Maturity function show in Equation 2.1.

N X M = Ij(T − T o)∆t (2.1) j=1

M maturity at age t expressed as TTF (time temperature factor) T average temperature of the concrete during time interval ∆t T o datum temperature ∆t time interval between readings

From the Nurse-Saul Maturity Function, the two major parameters are the temperature and the elapsed time after placement of concrete. Thus concrete temperatures should be monitored when water and cement start mix. This can also be related to Stage One of concrete hydration, however since it is very hard to monitor the concrete temperature while it is mixing. it has become standard practice to start monitoring concrete after it has been set in place. The Nurse-Saul Maturity function does have some limitations; Saul (1951) mentioned that this function is valid as long as concrete temperature did not reach about 50°C (122ºF) within the ﬁrst 2 hours or about 100°C (212ºF) within the ﬁrst 6 hours after the start of mixing. If the concrete temperature were to go above the recommended limit the maturity function could underestimate the strength.

2.4.4 Strength – Maturity Relationship

Once maturity index values are known several strength functions exist to determine concrete strength. A. Nykanen suggested an exponential strength-maturity function shown in Equation 2.2 (Nykanen, 1956).

kM S = S∞(1 − e ) (2.2)

Nykanen, states that the compressive strength is based on the water-cement ratio, and the constant k is linked to the initial rate of strength development over time and is dependent on the water- cement ratio and cement type.

12 S compressive strength S∞ limiting compressive strength M maturity index k a constant

Another popular strength-maturity relationship created by Plowman (1956) used a logarithmic equation shown in Equation 2.3 (Plowman, 1956).

S = a + blog(m) (2.3)

a strength for maturity index M=1 b slope of the line M maturity index

When this equation used on a log scale it plots a straight line showing the trend of the Maturity over time. However, this equation does have some ﬂaws. It does not provide a good representation for high and low maturity index values (Carino, 2001).

13 CHAPTER 3

METHODOLOGY

3.1 Introduction

In this chapter, three phases of the research will be covered. Phase one is about how the replacement slab was constructed and tested on a small scale and also goes into detail about how the mix design was acquired, tested and modiﬁed. Phase two covers how the research was expanded to a larger scale and puts focus on how industry operates with SCC and replacement slabs. Phase three will cover how to predict compressive strength of self-consolidating concrete using the steps outlined in ASTM Standard C1074-11.

3.2 Phase One

In phase one, two main sections will be covered; the ﬁrst phase will describe the development of the SCC mix and how it was developed and tested. The second section discuss the initial development of the replacement slab and how it was tested.

3.2.1 Development of the SCC Mix

The initial task of this research was to gather an SCC mix that met Florida Department of Transportation (FDOT) requirement that before the traffic lane can be opened up to vehicles, the concrete must reach a compressive strength of 2,200 psi in 6 hours. Besides FDOTs requirement, the SCC mix had to have specific characteristics such as high slump flow for rapid discharge and finish without segregation. To accomplish this a meeting was setup with a chemical company named BASF, headquartered in Ludwigshafen, Germany to discuss the appropriate admixture types and dosage rates that would achieve a combination of a high workability and high early strength SCC mix. The company representatives shared their experience in using 4x4 mix system (4000 psi in 4 hours) for accelerated slab replacement in California and provided technical information on the most effective admixtures and dosage rates for their SCC mixes. With this information, along

14 with speciﬁcations from FDOT and other useful material from professionals who have experience in Florida concrete mixes, an SCC mix was be developed for this project.

3.2.2 SCC Mix Ingredients

Table 3.1: Basic Mix Design, Initial Mix Design Without Any Modiﬁcations

Ingredients Units Admixtures Units Cement 830 lb Rheotec Z60 42 oz Coarse Aggregate 1625 lb Glenium 7700 33 oz Fine Aggregate 1244 lb Pozzolith 122 HE 515 oz Design Water 280 lb Pozzolith 700N 42 oz W/C 0.34 Air Content 2 %

Based on the specific SCC mix design goals of the project and industry input, a basic SCC mix design was developed for the trial batches, testing and further adjustment. The basic mix had to meet two objectives, high workability at placement and finishing, and early strength at lane opening to traffic. Table 3.1 shows the basic SCC mix design and Table 3.2 shows six trial mixes batched in this task. One of the goals of the basic mix design was to keep the amount of cement a low as possible, this is because when excessive cement is used the higher the chance of premature cracking (Kosmatka S, and Panarese, W 2002). To help keep the cement content to a minimum, a low water to cement ratio was chosen. The low w/c ratio also helps with high early strength (Kosmatka S, and Panarese, W 2002). The coarse and fine aggregates were sourced locally around Florida and passed FDOT specifications to be considered usable in FDOT concrete construction. For coarse aggregates FDOT requires testing the stone for gradation (AASHTO T-27), passing #200 sieve (FM 1-T011), Los Angeles Abrasion (FM 1-T011), Bulk Specific Gravity (FM 1-T085), and Acid Insoluble for FC (FM 5-510). The fine aggregates testing includes gradation (AASHTO T-27), passing #200 sieve (FM 1-T011), Bulk Specific Gravity (FM 1-T084), Fineness Modulus (AASHTO T-27), and color (AASHTO T-27). After the material was delivered gradation, specific gravity and absorption of coarse and fine aggregates was doubled checked and can be seen in Tables 3.3. The coarse aggregate is classified

15 Table 3.2: Six Trial Batches With Modiﬁcations

Ingredients Unit Trial Mix Trial Mix Trial Mix Trial Mix Trial Mix Trial Mix 1 2 3 4 5 6 Volume ft3 4.5 3 3 4 5 6 Cement lb 138 92 92 168 100 92 Coarse Aggregate lb 271 181 181 271 201 181 Fine Aggregate lb 204 136 136 204 147 136 Design Water lb 47 31 31 47 34 31 Batched Water lb - - - - 29 29 W/C 0.34 0.34 0.34 0.34 0.34 0.34 Air Content % 2 2 2 2 2 2

Admixtures Unit Rheotec Z60 oz 7 4.6 4.6 7 3 4.6 Glenium 7700 oz 6 3.7 3.7 6 3 5.5 Pozzolith 122HE oz 86 57 57 86 60 52 Pozzolith 700N oz 7 4.6 4.6 7 4 4.6

as grade 57 limestone with one inch maximum aggregate size. This aggregate size has proven to produce mixes suitable for long performing concrete pavements. In Florida, limestone is the main coarse aggregate used in concrete construction, which is the primary reason why it was used in this research. An absorption of 3.3% is in the medium range for Florida Limestone aggregates. The ﬁne aggregate was silica sand with a Fineness Modulus of 2.51. It should be noted that the source and properties of aggregates may change depending on the aggregate source and location in the state or from sources outside Florida. As such in future research, the mix design should be adjusted when the type and properties of aggregates are changed. Four admixtures were used in the mix, including a workability retainer, a high-range water- reducer, an accelerator and a set retarder/water reducer. Table 3.4 illustrates the types of admixtures and the description of their functions in the SCC mix. The dosage rates shown in Table 3.1 are in ﬂuid ounces per each 100 lb of cement, and they were calculated for the cubic yard and cubic feet of the mixes. Some of the admixtures had been successfully used in the production of the 4x4 mixes for accelerated pavement patching in California. The dosage rates of the admixtures were adjusted to

16 Table 3.3: Coarse and Fine Aggregate Properties, #57 Limestone and Silica Sand Produced by Vulcan

Coarse Fine Aggregate Aggregate Sieve Size % Passing Sieve Size % Passing 11/2(in) 100 #4 99 1.0(in) 96 #5 95 3/4(in) 71 #16 82 1/2(in) 32 #30 54 3/8(in) 14 #100 1 #4 2 #8 1 Fineness Modulus 7.1 2.51 Speciﬁc Gravity 2.53 2.63 Absorption 3.3% 0.4%

produce an SCC mix that meets the project criteria for slab replacement. The combination of the four admixtures was intended to provide the balance between high workability sustained during the discharge and ﬁnishing of the replacement slab, and the high early strength to meet the FDOT strength criteria for opening the section to traﬃc. The dosage where adjusted when changing the source/producer of the admixtures, construction environment and weather conditions during this research.

3.2.3 Trial Mixes

Six trial mixes were batched at the FSU-FAMU Civil Engineering Concrete Laboratory . The mixes were prepared in a 4 ft3 mixer, in a single or multiple batches to produce the required mix volume to perform the plastic concrete testing and to prepare required number samples for strength and other hardened phase testing. The volumes of the trial mixes ranged from 3 ft3 to 4.5 ft3. The aggregates were stored in covered bins to prevent ﬂuctuations in the moisture content. No adjustment was made in the mix water in trial mixes 1 to 4 as shown in Table 3.2. In trial mixes 5 and 6 the mix water was adjusted during batching to account for the water in the accelerator and for the moisture content of the coarse aggregate. The admixture dosage rates were adjusted to achieve a balance between workability and strength properties of the SCC mix. As shown in the test results,

17 Table 3.4: Admixture Descriptions

Admixture Description Rheotec Z60 Workability (slump) retaining admixture. Meets the interim requirements of ASTM C 494/C 494M Type S

Glenium 7700 High-range water-reducing admixture. Meets The requirements of ASTM C 494 Type A, water-reducing, and Type F, high-range water-reducing, admixtures

Pozzolith 122HE Accelerating admixture. Meets ASTM C 494 requirements for Type C, accelerating, and Type E, water-reducing and accelerating, admixtures

Pozzolith 700N Water-reducing and set-retarding admixture. Meets ASTM C 494/C 494M requirements for Type A, water-reducing, Type B, retarding, and Type D, water-reducing and retarding

Trial mix 5 and Trial mix 6 met the high workability aspect of the fresh concrete and the 6-hour strength requirement of the FDOT. It should be noted that when admixture source is changed, the dosage rates may have to be adjusted to meet the required SCC properties. Adjustments would be needed in the dosage rates of admixtures to account for transportation time to the jobsite.

3.2.4 SCC Testing – Plastic Phase

Tests on fresh SCC samples were performed on the trial mixes to evaluate the workability and suitability of the mix for rapid discharge, spread and ﬁnish. The test methods, brief descriptions, are shown in Table 3.5. The slump ﬂow and T20 (ASTM C1611) tests were performed on all trial mixes while the J-ring test was performed on some of the mixes. The J-ring test (ASTM C1621) is normally performed to determine the concrete’s passing ability through obstructions and reinforcing bars. It may not be relevant to SCC mix placement in large pits such as the replacement slab pits since the replacement slabs do not have reinforcement. However, this test was included because it is standard practice for testing concrete mixes. The column segregation test was performed only on Test Mix 2 (TM2). This test may not be relevant to slabs with relatively shallow depths and wide surface dimensions such as the replacement pits. However, the test provides a means to

18 reasonably determine the potential static segregation of self-consolidating concrete, but it was not found relevant to this research because the slab thickness was only 8 in. The acceptance criteria for the properties of conventional SCC mixes are shown in Table 3.6. It should be noted that the acceptance criteria pertain to workability of SCC mixes in the structural applications in heavily reinforced structural members and may not necessarily be relevant to replacement slabs that are relatively thin with wide surface dimensions. However, Table 3.6 was used as a guide to assess the suitability of the SCC mix design for slab replacement. More speciﬁc plastic concrete requirements will be proposed in subsequent chapters of this report. Except for TM3, which was prepared and tested at low temperature, the slump ﬂow measurements of the trial mixes used in this task ranged from 21 to 24 inch and the T20 ranged from 3 to 7 seconds.

3.2.5 Development of Precast Replacement Slabs

Development of Precast Replacement Slabs. The precast replacement slabs were designed to be a temporary concrete slab that would be installed before the SCC mix is poured. The initial precast slabs were designed to be 6ft square at 9in deep with two layers of rebar reinforcement. The grid of rebar was set at 1ft allowing a minimum or 1.5-inch cover. The panels were designed to withstand multiple field applications. The bottom of each panel included four 8in x ¾ in recessed stripes along the bottom of the panels, as shown in Figure 3.1. The stripes were positioned along the long side of the panel bottom in order to be perpendicular to the direction of traffic. The recessed stripes were designed to improve the friction and interlock with the base to minimize panel shifting under the action of the moving traffic. The perimeter was molded so that a greased gasket could be installed. The gasket served the purpose of protecting the edges of the concrete and the gasket can be seen below in Figure 3.1. The panels were casted with a 4000-psi mix from a local concrete batching yard.

3.2.6 Testing of Precast Replacement Slabs

To properly test the precast panels, a testing pit was dug to match the precast slab. The pit had a base layer of construction sand and the sides were made of concrete to simulate concrete that the precast panel might be against. A crane was used to pick up the precast panel and was swung into place. A picture of this can be see in Figure 3.1. After the panel was installed, it was tested for vertical and horizontal movement using a 11,000 lb forklift. The horizontal load was induced on

19 Table 3.5: Standard ASTM Tests for Fresh Concrete Properties

Test Description Slump Flow test determines the slump flow of self-consolidating concrete in the laboratory or the field. It is used to monitor the con- sistency of fresh, unhardened self-consolidating concrete and its unconfined flow potential. This test method is considered applicable to self- Slump Flow consolidating concrete having coarse aggregate (in) ASTM up to 25 mm [1 in.] in size. The rate at which C1611 the concrete spreads is related to its viscosity. Appendix X1 provides a non-mandatory procedure that may be used to provide an indication of relative viscosity of self-consolidating concrete mixtures. T20 (T50) test is a measure of the concrete’s viscosity and is measured as the amount of time it T20 (T50) (in) takes for concrete in the slump flow test to reach ASTM C 1611 a diameter of 20 in (or 50 cm) centimeters). J-Ring test is a measure of the concrete’s passing ability through obstructions and reinforcing bars. ASTM C 1621, ”Passing Ability of Self- Consolidating Concrete by J-Ring.” The J-Ring J-Ring ASTM is a cage of rebar that is set up around the slump C1621 cone. The slump flow test is run both with and without the J-Ring in place and the passing ability is the difference in slump flow. Column Segregation test provides a procedure to determine the potential static segregation of self-consolidating concrete. The test is used to develop self-consolidating concrete mixtures with segregation not exceeding specified limits. Column The static segregation of SCC is determined by Segregation % measuring the coarse aggregate content in the ASTM C1610 top and bottom portions of a cylindrical specimen (or column).The degree of segregation can indicate if a mixture is suitable for the application.

20 Table 3.6: General Acceptance Criteria for SCC Workability

Test Description Appropriate for members with light or no reinforcement, short lateral ﬂow 21-24 distance, or high placement energy (e.g. panels, barriers, Slump Flow (in) coping)

24-27 Ideal for most applications

long lateral ﬂow distance, or 27-30 low placement energy (e.g U- beams, I-beams, other beams) <2 Poor Stability

T20 (s) Acceptable, should not vary 2-7 over range of 3 s between batches

Possible, may reduce pal- >7 ceability Appropriate for members with highly congested re- <0.5 inforecement (e.g U-beams, I-beams, other beams) J-Ring (in) Appropriate for memebrs 0.5-1.0 with moderatly congested reinforcement

Appropriate for unrinforced >1.0 or highly reinforced members (e.g panels, coping) <5 Highly segregation resistant

Column 5-10 Segregation resistant Segregation (%) Borderline segregation resis- 10-15 tant

>15 Not segregation resistant

21 Figure 3.1: Precast Panel with Test Pit the slab by applying a sudden brake on the forklift. The differences in the horizontal displacement at different speed levels readings were not of significant values. The vertical displacements at the edge of the slab increased as the applied loads approached the LVDTs. The vertical and horizontal displacements of the slabs were measured using two vertical and two horizontal LVDTs (Figure 3.2).

Figure 3.2: Field Test on the Precast Panels

22 3.3 Phase Two

Phase two of the research explains the methods used to upscale the size of the SCC mixes, precast replacement slabs, and concrete maturity curves. Increasing the size of everything brings the research closer to real world applications, the same methods from phase one were used and adjusted them to ﬁt the larger scale and test the on a section of test track.

3.3.1 Location

A test track in an industrial park was used for this research in Green Cove Springs, Florida. The site also includes a concrete recycling plant. The oval shaped test track is 400 ft long and 14 ft wide asphalt pavement, as shown in Figure 3.3. The test track had been used to evaluate pavement products and innovative technologies. Traﬃc loads are simulated using a loaded truck traveling at 10 mph around the track.

Figure 3.3: Pavement Test Track

3.3.2 Preparation of Replacement-Slab Test-Pit

A 24ft x 14ft section of the test track pavement was removed to construct the replacement-slab test-pit and the surrounding concrete pavement. After removing the asphalt and base layers, forms were erected to setup the test pit as shown in Figure 3.4. The ﬁnal pit dimensions were 12ft 3in x 12ft 3in x 8in and was designed to accommodate two 6ft x 12ft x 8in precast panels. To simulate

23 an existing concrete pavement, a 4,000-psi concrete mix was prepared in a commercial batch plant and cast around the test pit as shown in Figure 3.5. The concrete was cured using a plastic sheet.

Figure 3.4: Forms For the Replacement-Slab Test-Pit

After construction of the test pit, a layer of fine recycled concrete aggregate was placed at the bottom of the pit and compacted as a leveling course to establish the design pit depth of 8” as shown in Figure 3.6. Because of its stiffness, fine recycled concrete aggregate was found to be a suitable leveling course to establish the required depth for the slab replacement pit. Other leveling material options would include builder’s sand or geosynthetic mats. Builder’s sand was used at FSU site to construct the 6ft x 6ft x 9in test pit and was described in phase one. In general, the leveling course is necessary if the contractor uses a single-thickness precast slabs in pits with varying depths. No dowel bar holes were drilled in the two faces of the adjacent pavement. The high cost of drilling and the difficulty in finding a contractor for such a small job were the reasons for omitting this step. It should be emphasized that this step is part of the precast/SCC system being proposed. This activity would normally involve drilling the holes for the retrofit dowels, removing debris and fines from the hole, and covering the hole with tape to prevent contamination during the installation and removal of the temporary precast panels. Prior to casting concrete for the new replacement

24 Figure 3.5: Paving Concrete Around the Replacement Slabs Test Pit slab, the tape would be removed, the drilled holes would be ﬁlled with epoxy and then the new dowel bars would be inserted.

Figure 3.6: Test Pit With Fine Recycled Concrete Aggregate as Leveling Course

The selection of the slab replacement dimensions for the pit was made based on the fact that a 12ft x 12ft is the maximum design template specified in the proposed precast/SCC system. These dimensions were also specified in FDOT Index 308. Other templates specified in the proposed method for slab replacement, include 6ft x 12ft, 8ft x 12ft , and 10ft x 12ft. These templates require the contractor to prepare only two groups of precast panels with 6ft x 12ft and 4ft x 12ft

25 dimensions. These two sizes can ﬁt individually or in combination to ﬁt all the proposed pit sizes, namely, 6ft x 12ft, 8ft x 12ft, 10ft x 12ft, or 12ft x 12ft, in slab replacement projects as shown in Figure 3.7. The dimensions of the test pit in Green Grove Spring site allowed the installation of two standard 6ft x 12ft panels (Figure3.7) with the 12ft sides being in the transverse direction. Once installed,

Figure 3.7: Temporary Precast Panels for 6’ x 12’, 8’ x 12’, 10’x 12’ and 12’x 12’ Replacement Pits three joints were in the direction of traﬃc. Two of the joints were bordering the surrounding concrete, which represented an existing pavement structure, and the third joint was between the two panels. This arrangement was chosen to test the structural stability of the two precast panels

26 installed side by side under simulated traffic loading conditions. This installation pattern was considered the most venerable case scenario since the middle joint would be between the two movable panels, compared to a single panel (Figure3.7a) installed with its two transverse joints firmly pressed against sides of the fixed pavement. It should be noted that the edges of the temporary panels would not be doweled or structurally connected to the surrounding concrete.

3.3.3 Construction of Temporary Precast Panels

The two precast panels were fabricated on site at the test track. The panel dimensions were 6ft x 12ft x 8in. These planar dimensions were approximately 3in shorter than the pit dimensions, which would create a 1.5in gap between each panel side and the surrounding concrete. This was to facilitate the installation and removal of the panels from the test pit. It should be noted that a contractor may choose another option, which is to reduce the dimensions of the precast panels by 1in to 3in and maintain round numbers for tests pit dimensions (e.g. 6ft x 12ft, 8ft x 12ft, 10ft x 12ft, or 12ft x 12ft). The detail design of the two panels is shown in Figure 3.8. The preparation

Figure 3.8: Precast Slab Design of forms, the assembly of the steel reinforcement and installation of the four lift anchors were completed on site as shown in Figure 3.8. The inside perimeter of the forms was ﬁtted with a

27 plywood strip to create a 1.5in wide and 1.5in deep recess in the middle top half of the slab to contain the foam gasket around the panel. Also, ﬁve plywood boards 8in x 3/4in were fastened to the bottom of the form to produce four recessed strips along the bottom of the panels in Figures 3.8 and 3.3.7. The forms were reinforced in two layers using #4 reinforcing bars in the ﬁrst panel and

Figure 3.9: Lift Anchors in the Precast Panels

#5 bars in the second panel. The utilization of the two bar sizes was intended to determine the structural adequacy of the panels under repeated traffic load and to optimize the final design and panel weights. The reinforcing bars were spaced at 12” on centers in both directions. Also, four lift anchors model were embedded in each slab and were fastened to the reinforcement. The recess of each anchor was secured with a disposable plastic cover, which could be removed and reused at the time of slab rigging (Figure 3.10). Covering each recess after installing the panels in the pavement pit is necessary to protect and protect the anchors when panels are under traffic and to eliminate pavement noise. The panels were cast at the test track using 4,000 psi concrete supplied by a commercial concrete plant (Figure 3.11). The concrete was vibrated with emphasis around the lift anchors and the recessed components of the form to ensure that the panels would have firm lift anchors and perfectly shaped recessed areas. After casting, the forms were covered with plastic sheets for proper curing. The panels were designed to be robust to withstand multiple field applications. The bottom of each panel included four 8in x ¾in recessed strips along the bottom of the panels, as shown in Figure 3.12. The strips were positioned along the long side of the panel bottom in order to be perpendicular to the direction of traffic. The recessed strips were designed to improve the friction and interlock with the base to minimize panel shifting under the action of the moving traffic.

28 Figure 3.10: Locations of Covered Lift Anchors

Figure 3.11: Casting of Precast Panel

29 The backer rod was also intended to absorb minor impact of the panel sides and corners against the surrounding concrete, and thus preserve the structural integrity of the panels. Another possible beneﬁt may be its joint sealing ability to minimize ingress of surface water during the temporary use of the panels in the replacement pits. The protruding portion of the backer rod was lubricated

Figure 3.12: Striped Panel Bottom and Backer Glued Baker Rod Along the Panel Sides with polymer gel to facilitate insertion and removal of the panel in and out of the slab replacement pits. The polymer gel is degradable and environmentally friendly lubricant (Figure 3.13). The lubricant that was prepared by adding 0.35 oz of concentrated synthetic polyacrylamide polymer to 8 liters of water. The lubricant was prepared 24 hours before its application to allow time for the viscosity of the gel to increase with time. The ﬁnal product was a viscous water-based gel that could be used safely to reduce labor/material cost of replacing the backer rod every time the precast panel is used.

Figure 3.13: Lubricated Backer Rod

30 3.3.4 Installation of Precast Panels

The two precast slabs were cured until reaching the required concrete strength. After stripping oﬀ the panel side forms, installing the backer rod and removing the plastic covers from the lifting anchor recesses, the lift rings were clutched to the anchors. Using steel chain, each panel was lifted from the base form, as shown in Figure 3.13. An excavator was used to lift the panels and to install them in the test pit, as shown in Figure 3.14. The choice of using the excavator was due to its availability nearby at the recycling plant. On job sites, the contractor may choose this or other lifting tractors or backhoes. However, it is recommended to use a more eﬃcient and steady lifting

Figure 3.14: Rigging and Maneuvering the Precast Panel system of a backhoe/tractor/forklift with a steel I-beam attachment, as shown in Figure 3.15. The service life of the precast panels is related to the methods and equipment used in storing, transporting, delivering, and installing of the panels. The precast panels can easily be damaged if they are mishandled at any stage before or after their use. Figure 3.15 shows three diﬀerent proposed rigging methods, which can be used to maneuver the precast panels in the ﬁeld. These methods are as follows: 1. Use of a four-sling-lift

2. Use of an equalizer beam

3. Use of a spreader frame

The ﬁrst method (Figure 3.15a) provides more ﬂexibility to adjust and rotate the panel while lowing it in the pit. This method was used to rig the precast panel at FSU test site, and it was found to be

31 Figure 3.15: Alternative Rigging Methods suitable to maneuver a 6ft x 6ft panel (Figure 3.16). However, the disadvantage of this method is the constant horizontal adjustment that is required of a larger precast panel when lowering it in the pit. The equalizer beam and the spreader frame methods provide better stability (Figure 3.15b and c) in the rigging process, but the ﬂexibility of rotating the panels is more restricted in these cases. Prior to installation in the test pit at the test track, the panels went through multiple lifting and

Figure 3.16: Use of a Four-Sling-Lift Method and a Boom Crane at FAMU-FSU Test Site lowering cycles including. These maneuvers were performed to test the robustness of the anchors, speed of the handling method, and the adequacy of the structural design of the panels. After completing multiple handling and moving maneuvers, no cracks or damages were observed in both panels. The two panels were then lowered and positioned inside the replacement test pit as shown in Figure 3.17. The process of lowering each panel took about 10 minutes to complete. A faulting of about ¼in was measured at the joints between panels and with the surrounding pavement. This faulting may be considered acceptable since the precast panels are temporary ﬁllers in the

32 slab replacement pits with installation period not exceeding one week. At job sites and with the

Figure 3.17: Placement of the Two Precast Panels proper equipment, adequate tools, and practical experience, the process of installing the panels in slab replacement pits would not require more than 10 minutes. In addition, it should be noted that proper leveling of the surface of the soil underneath the precast panels is important to establish a uniform contact area and to prevent any “rocking” action of the panels. A proper leveling tool would be recommended to ensure uniformly leveled bottom of the pit. The leveling course, compacted by the weight of the panels and traﬃc, should remain in place prior to casting the replacement slab.

3.3.5 Instrumentation and Test Loading the Panels

With the two precast panels installed in the test pit, eight LVDTs were placed at key locations on the surface of the panels as shown in Figure 3.18. The LVDTs were intended to measure panel movements under the truck wheel loads, and during the braking maneuver of the truck on the panels. The movements of the panels were monitored in the vertical and horizontal directions and were recorded using a data acquisition system controlled by a laptop computer. Each LVDT was instantaneously monitored using external readout units placed on the two sides of the test section to obtain real time displacement readings during the load testing. A 60,000 lb. concrete pump truck was used to load the precast panels as shown in Figure 3.19. The wheel conﬁguration included a single and a double tandem axle. According to the truck speciﬁcations, the front axle carried ¼

33 Figure 3.18: Instrumentation of the Precast Panels at the Test Track

(15,000 lb) of the total weight and the rear axles supported the remaining ¾ (45,000 lb) of the truck weight. This truck represents one of the heaviest on urban streets. The truck made 50 passes on the panels. It traveled at a speed lower than 10 mph on the precast panels. The movements of the panels were being observed, and the vertical and horizontal displacement measurements were being measured by the LVDTs. During the initial passes, some seating of the panels was noticed. However, upon subsequent passes, no further seating was detected. The real time measurements from the readout units were helpful to determine which corners of the two slabs were moving more than the other corners. The LVDTs readings showed that the upper right corner of the second slab was displacing slightly in vertical direction more than the other corners. In general, the panels did not rock under the weight of the wheels. This was an indication of a stable support generated by the striped bottom configuration of the panels and the effect of their initial panel seating in the leveling course. The truck made additional passes traveling at 10 mph and then suddenly braking on the panels. This braking maneuver was intended to verify whether the panels were shifting horizontally as the truck suddenly braked on the panels. However, no noticeable shifting was detected. This was an indication of friction and interlocking action between the stripes bottom of the panels at the leveling course. The LVDTs were simultaneously capturing the vertical and horizontal displacements. The initial readings from the LVDTs confirmed the observations of some minor and insignificant movements in the horizontal and vertical directions after several loading cycles.

34 Figure 3.19: Front and Rear View of the Truck Used in Load Testing of Precast Panels

3.3.6 Removal of Precast Panels from the Test Pit

The precast panels were left in the test pit for a period of four weeks prior to their removal. A day prior to casting the replacement slab, the two panels were removed using the same excavator used in the original installation. The removal of both panels took less than 10 minutes. However, during lifting of the precast panels, the corner of one of the panels was chipped off as a result of an impact with the surrounding concrete (Figure 3.20). With proper equipment and experience, the removal of panels from the slab replacement pits should be accomplished in less than five minutes without any damage to the sides of the panels or to the surrounding pavement. This experience brought attention to the need to use better and more effective handling methods, since the panels would be re-used multiple times in the rehabilitation projects.

3.3.7 SCC Mix - Field Evaluation

The efforts prior to this task were focused on preparing an SCC mix with high workability, flow rate and high early strength. These results were achieved and verified at the materials laboratory of FAMU-FSU College of Engineering. The final mix design is shown in Table 3.7. All mixes were batched in the lab using a 4ft3 concrete mixer. The workability of each mix was tested immediately after adding and thoroughly mixing the accelerator using the slump flow and T-20 test methods according to ASTM C1611/C1611M-14. The 12ft x 12ft x 8in replacement slab at the test track pit required the preparation of 4 cu yd of SCC mixed at a batch plant and transported to the job site.

35 Table 3.7: SCC Final Mix Design

Ingredients lb/yd3 Description Type I/II cement used – Cement 830 AASHTO M85

Coarse Aggre- 1625 Limestone grade size 57 gate

Fine Ag- 1224 Silica sand gregate

Mix water adjusted to aggre- Mix Wa- gate moisture and water con- 280 ter tent in the accelerator admixture

W/C 0.34

Admixtures ﬂ oz/yd3 Description ASTM C 494/C 494M, Type Workability 42 S, speciﬁc performance ad- Retainer mixture

ASTM C 494, Type A, water- HRWR - reducing, and Type F, high- Polycar- 50 range water-reducing admix- boxylate ture

ASTM C 494 Type C, accelerating, and Type E, water- re- Accelerator 473 ducing and accelerating, and admixture.

ASTM C 494/C 494M, Type Water Re- A, water-reducing, Type B, ducer & 42 retarding, and Type D, water- Retarder reducing and retarding admixture

36 Figure 3.20: Damage to One of the Precast Panels During Removal

It was noted that the conditions were quite different when batching the SCC mix at a batching plant as compared to laboratory conditions. The main difference was the large volume of the mix (4 cu yd) batched at the commercial batch concrete plant in a 10 cu yd. truck mixer compared to batching at a 4 cu ft in a university laboratory. Also, transportation time between the batch plant and test track was estimated between 45 to 60 minutes. This would affect the SCC workability. Also, upon arrival to the jobsite the accelerator would be added which requires special handling to the truck mixer to avoid quick setting of the concrete inside the mixer before discharging the entire load of concrete. The primary issues that had to be addressed in batching the SSC mix at the concrete plant were the following:

1. Maintaining workability over an extended period following the mixing of the accelerator in the SCC mix.

2. Impact of changing material sources.

3. Diﬀerences in moisture conditioning of materials at the concrete plant compared to FAMU- FSU laboratory with respect to SCC plastic and hardened properties.

4. Determining the extent of workability loss in the SCC mix during the transportation period from the plant to the jobsite.

5. Extent of workability retention after addition of the accelerator at the jobsite.

37 3.3.8 Veriﬁcation of Workability Retention at FAMU-FSU Laboratory

To address the above issues, tests on freshly mixed SCC were performed to evaluate workability of the mix over a period of 60 minutes after mixing the accelerator. During slab replacement at job sites the SCC mix must be transported from the batch plant to casting place. Transportation requires considerable amount of time ranging from 45 to 60 minutes. At the job site, the accelerator is usually added to the truck mixer, followed by additional mixing before casting the slabs. After casting the first slab, the truck mixer will be idle for a few minutes and continue revolution in an agitation mode while traveling to the next slab. To simulate the impact on the SCC workability because of truck idle time, it was decided to evaluate the slump flow of the mix for 60 min after addition of the accelerator. To verify the workability retention, laboratory tests were performed on two batches B1 and B2 using the same mix design in Table 3.7. The evaluation involved performing the slump flow and T-20 tests at different time intervals over a period of 60 minutes. After completing the batching sequence and performing the initial slump flow test on the SCC mix, the speed of the revolving mixer was reduced to simulate a case of an idle truck mixer in an agitation mode. Also, the mouth of the mixer was covered with a plastic sheet to prevent evaporation of the mix water. Concrete temperature inside the mixer was regularly monitored between the slump flow tests using an infrared thermometer device as shown in Figure 3.21. This was to determine if the addition of the accelerator would cause an increase in cement hydration while the concrete mix was in the agitation mode. A significant increase in concrete temperature would have indicated a high level of hydration which would lead to loss of the concrete workability. The concrete temperature during this test period increased by only 2 degrees Fahrenheit. Argos Ready Mix Company provided support to the research team by offering to supply the concrete for the replacement slab at the test track. The research team and Argos staff conducted additional trial batches of the SCC mix at their concrete plant in Jacksonville. The aim was to repeat the workability retention performance of the SCC mix using materials from the Jacksonville area, before batching 4.5 cu. yd mix quantity needed to cast the replacement slab at the test track. Using the SCC mix shown in Table 3.7, five 1.25 cu ft trial batches were prepared and tested using the Argos concrete plant materials, equipment and testing lab. The five batches were neces-

38 Figure 3.21: Temperature Measurement of SCC Mix sary to prepare a sufficient quantity of concrete for the multiple slump flow tests and cast cylinders for subsequent compressive strength tests. The same batching sequence was followed in all five trial batches, using the same aggregates and cement. The as-batched water content had to be changed for the different batches, since the free moisture of the sprinkled aggregate-pile changed during the four-hour batching period. In addition, fine white material was present with the coarse aggregate obtained from the interior of the pile, which retained more free moisture compared to the free moisture content measured from pile specimen. It was noticed that the fine aggregate in Jacksonville plant had different specific gravity than the aggregates used in the FAMU-FSU trial batches. Variability in material moisture condition is to be expected during a production day at concrete plants. This presented a unique opportunity to test the flexibility of the designated SCC mix in achieving the same range of workability and early strength values at a production plant compared to a small mixer in a controlled environment of a university laboratory.

3.3.9 Casting of Replacement Slab in Test Pit at Test Track

With the conﬁdence generated in the SCC mix during the trial batches at the Argos plant, it was decided to batch a 4.5 cu yd mix at the closest concrete plant to the test track using the SCC mix shown in Table 3.7. All the ingredients and the admixtures except for the accelerator were batched in the same sequence of the trial batches at the plant. The mix water was adjusted to account for the free moisture in the plant aggregate. Also, a certain quantity of the mix water was

39 withheld to partially account for the water content of the accelerator at the casting site, and to retain suﬃcient quantity in case the mix required more water to achieve the required workability at the site. The results and analysis of the SCC pour into the test track is discussed in Chapter 4.

3.4 Phase Three

Phase three explains the method used to create a maturity curve for the SCC mix. A maturity curve accurately predicts compressive strength of the self-consolidating concrete mix (Table 3.8) at a given time after it has been poured. The following sections will describe the methods used to achieve the maturity curve.

3.4.1 Location and Test Setup

All tests conducted were located at the FAMU-FSU college of engineering materials laboratory. All test procedures were conducted based on the ASTM C 1074-11 (Standard Practice for Estimat- ing Concrete Strength by the Maturity Method). The curing process of the prepared specimens was in accordance to ASTM C31. Two mixes were completed separately and a total of 44, 4in by 8in cylinders were casted per mix. These cylinders had designated times to be tested starting 2 hours after the concrete was placed and ended at 28 days later. All cylinders were tested according to ASTM C39/C39M – 12a and were capped using sulfur capping compound from Humboldt Manufacturing. The difference between the two mixes was the source of the materials. The first SCC mix was developed using Argos materials in Tallahassee and Jacksonville. The second mix used materials from Anderson- Materials (A-Materials) in Tallahassee. Although both concrete suppliers provided cement type I/II, the hydration rates of the two cements were different. According to A-Materials, their cement hydrates faster than the same type from Argos. Also, the two fine aggregates had different specific gravities. Additionally, A-Materials coarse aggregate was noticed to contain more fines than Argos. Therefore, it was decided to adjust the mix design to for the material acquired from A-Materials shown in Table 4.10. The final batch needed to fill the required number of test cylinders was found to be 3 cubic feet.

40 Table 3.8: Final SCC Mix Design Argos Versus A-Materials

Argos A-Materials Ingredients Mix lb/yd3 Mix lb/yd3 Cement 830 830 Coarse Aggregate 1625 1625 Fine Aggregate 1224 1157 Mix Water 280 280 W/C 0.34 0.34 Admixtures ﬂ oz/yd3 ﬂ oz/yd3 Workability Retainer 42 42 HRWR - Polycarboxylate 50 50 Accelerator 473 473 Water Reducer & Retarder 42 42

3.4.2 Instrumentation

Several diﬀerent devices were used to measure temperatures, according to ASTM C1074-11 any device that can monitor and record concrete temperature is acceptable. Besides using the

Figure 3.22: Temperature Reading Devices (Tawﬁq and Armaghani, 2016) recommend commercially available maturity meter, other devices such as thermal cameras and infrared temperature reading guns were used (Figure 3.22). These devices make it possible to have a novel approach to establish a maturity curve.

41 3.4.3 Concrete Pouring

Using the mix designs found in Table 4.10, five total concrete mixes were performed. A five cubic foot mixer was used for batching. All batching had the same sequence of adding ingredients in the mixer. The first step was to wash out the mixer with water. Then half of the course aggregate was added followed by half the fine aggregate this process was repeated until all the course and fine aggregates were combined. Then three of the admixtures were combined with the batch water. The batch water/admixture was then added to the mix. The accelerator (Pozzolith 122HE) was held until all water and cement have been added. Once all the ingredients have been mixed a slump was taken and recorded. The first mix was a practice mix; a trial to see how to setup and run all the thermal devices. To accurately measure the temperature on all devices within a minute of each other it was recommend having least two persons recording the temperatures. The second concrete mix used Argos materials, 44 (4in x 8in) concrete cylinders were casted. Two of the samples had thermal couples embedded in them and were used for 28-day testing. The third mix used A-Materials products that included their course/fine aggregates and cement. Mix three followed the same procedure as mix two with the same number of casted cylinders. The final two mixes were performed on the same day. The purpose of these two mixes was to either confirm or deny the maturity curves developed from second and third mix. The two mixes were poured 30 minutes apart to give adequate time to collect and test data.

3.4.4 Concrete Curing

All the concrete casted in this part of the research was cured in the lab. The concrete was not exposed to direct sunlight after being casted and all samples were sealed with plastic caps or plastic sheeting. This caps and sheeting provided a barrier so that water could not evaporate away from the concrete, but instead accumulate on the plastic or concrete. A water bath was not used in this project, since most of the cylinders were crushed before 24 hours had elapsed. However, all cylinders remained sealed until they were tested.

3.4.5 Cylinder Testing

The cylinders were tested using a Forney concrete compression machine that had a maximum capacity of 400k pounds. As mentioned before, the cylinders were test every hour starting at 2

42 hours after the pour time. However, when the testing had begun, it was quickly realized that the concrete had not gained enough strength at the 2 hour post pour time and the testing started at the 3 hour mark instead. On every hour before the cylinder was crushed, the temperature was taken and the cylinders that were being tested had to be quickly unsealed, dried, sulfur capped and then placed in the machine. For the ﬁnal two mixes that checked the maturity curves, the concrete cylinders were not tested every hour but instead were tested when the maturity curves indicated certain strength. The theoretical strength was then compared to the results from the concrete that was tested.

43 CHAPTER 4

RESULTS AND ANALYSIS

4.1 Introduction

In this chapter, results of every stage of this project will be displayed and discussed. The first section will cover the fresh and hardened concrete properties of the SCC mixes that were performed at the Tallahassee, FL location. Similar to the Phase One discussed in Chapter 3, the results of the precast slab testing will be discussed next. Phase Two will discus the fresh and hardened properties of the final SCC mix and the final precast slabs used. The development of the maturity curve for the SCC mix will be described in Phase Three.

4.2 Phase One Results

Phase one of the results will be broken down into three sections; ﬁrst being fresh concrete properties of the trial mixes, then it is followed by a section that’s dedicated to the hardened properties. The last section covers the results from the initial replacement slab tests.

4.2.1 Fresh Concrete Properties

The trial mixes were batched at different ambient temperatures as shown in Table 4.1. Ad- justments were made in the admixture dosage rates at high or low temperatures to maintain high workability and achieve the required lane-opening compressive strength. For example, the quantity of the accelerator could be reduced when the ambient temperature exceeds 85 degrees Fahrenheit and may need to be increased when the ambient temperature is lower than 50 degrees Fahrenheit. Some adjustment to the cement content was also considered. The slump flow and T20 tests measure the extent of spread and velocity of the flow of the SCC at discharge. The results for these tests as shown in Table 4.1 indicated a very good to moderate spread properties of the designed SCC mix for slab replacement, after addition and mixing of the accelerator admixture. These results suggest that the designed SCC mix can be discharged at a high rate to rapidly fill the open pit of the replacement slab. This will result in higher production

44 Table 4.1: Ambient Temperatures and Fresh Concrete Test Results

Tial Ambient Slump T20 J-Ring Column Mix Temperature Flow Segregation °F in seconds in % TM1 84 23.3 3 >1 –

TM2 79 21 4 >1 7.5

TM3 57 16 >7 >1 –

TM4 82 22 5 – –

TM5 83 24 4 0.9 –

TM6 83 21 6 – –

rates. Except for TM3, small variations in the ambient temperature seems to have minor, if any, impact on the workability of the SCC mix as evident from the slump ﬂow and T20 test results.

4.2.2 Hardened Concrete Properties

Compressive strength samples were prepared from each trial mix with the exception of TM2, where only fresh SCC tests were perfromed. A set of three 6-inch diameter cylinders was prepared for each testing age. Table 4.2 shows the age of tested samples. In preparing the cylinder samples, the molds were filled with the SCC mix without the use of any vibration or rodding. Instead the concrete scoops full of concrete were droped from a hight 2 inches above the cylinder rim. This was a departure from traditional sample preparation for compressive strength. The test results in Table 4.2 represent compressive strength of cylinder samples prepared with no vibration or rodding. The strength test results are also illustrated in Figure 4.2. To determine the impact of lack of vibration or rodding on strength of the concrete cylinders, companion sets of vibrated and non-vibrated samples were prepared from Trial Mix 1 (TM1) (Table 4.3) . A vibrating table was used to consolidate one set of the samples. The other set of samples were filled without vibration. The cylinders were weighed and tested for compressive strength at 1 3, 7, and 28 days. The test results and percent variation in strength are shown in Table 4.2 and are illustrated in Figure 4.2. The test results show little or no difference in the weight or the strength values between the vibrated

45 Figure 4.1: Temperature Measurement of SCC Mix and non-vibrated samples. Based on these test results, samples from the remaining SCC trial mixes were prepared without vibration. It should be noted that no segregation was observed in any of the vibrated or non-vibrated samples. In fact the result of the column segregation test performed on TM2 (Table 4.1) confirmed that the mix was “segregation resistant” as described in ASTM C1610. The compressive strength at early ages (Table 4.2) seems to be impacted by the ambient temperature as shown in TM3 compared to the other trial mixes. The FDOT’s 6-hour stength of 2,200 psi was achived in TM5 after adjusting the dosge rates of the four admixtures. By reducing the amount of the high-range water reducer in TM5, while maintaining the rate of the accelerator constant, and with the help of higher ambient temperature, a strength of 4,203 psi was achieved at 6 hours, which is significantly higher than the required lane-opening strength of 2,200 psi. The team decided to lower the 6-hour strength by preparing and testing TM6. In TM6 the accelerating admixture Pozzilith 122HE was reduced and the HRWR (Glenium 7700) was increased. This change in the mix ingredients reduced the 6-hour strength to 2,752 psi and maintained a good slump flow of 21and aT20 of of 6 seconds.

46 Table 4.2: Compressive Strength Test Results (psi)

Trial Ambient Temp. 4 6 9 12 24 3 7 14 28 Batch °F hrs hrs hrs hrs hrs days days days days TM1 84 – 1788 – – 4879 6707 7682 8229 8600 TM2 79 Only Fresh Concrete Properties Tested TM3 57 270 549 1243 2044 3290 – – – – TM4 82 143 1808 3215 4369 6606 – – – – TM5 83 2439 4230 5438 6197 7055 – – – – TM6 83 890 2751 4322* – – – – – – *Testing At 8 hrs

Figure 4.2: Compressive Strength of Vibrated and Non-Vibrated Cylinder Samples

Table 4.3: Compressive Strengths and Weights of Vibrated vs. Non-Vibrated Samples of TM1

1 Day 3 Days 7 Days 28 Days Sample Compaction Method Wt. Fc’ Wt. Fc’ Wt. Fc’ Wt. Fc’ (lb) (psi) (lb) (psi) (lb) (psi) (lb) (psi) Vibrated 28.7 4862 28.7 6905 28.6 7623 28.8 8458 Non-vibrated 28.6 4869 28.5 6312 28.0 7799 28.3 8882 % Diﬀerence (NV/V) x 100 99.6% 100% 99.3% 91% 98% 102% 98% 105%

47 4.2.3 Precast Panel Test Results

Figure 4.3: Vertical Displacement of Precast Panels

Testing of the Precast Panels used strategically placed LVDTs that measured vertical and horizontal displacement. When loading the panel with the rear wheels of a 11,000 lb forklift and moving close to the edge of the slab where the LVDTs were located. The vertical displacements increased to about 0.04 to 0.06 inches. Some spikes in the LVDTs reading were noticed, and these peaks were due to the vibration in the slab from the moving forklift. The duration of these peaks was a fraction of a second. Because the loads from the rear wheels were not precisely centered between the vertical LVDTs as seen in Figure 4.3. A similar pattern of displacements was noticed from the horizontal LDVTs (Figure 4.4). For the horizontal displacement, the precast panels showed a maximum value of about 0.005 in.

4.3 Phase Two Results

In this section the results of Phase Two will be broken down into three parts. The ﬁrst will describe the fresh concrete properties of the ﬁnal SCC mix design along with the hardened concrete

48 Figure 4.4: Horizontal Displacement of Precast Panels properties. The ﬁnal part will analysis the displacements recorded of the ﬁnal precast panels installed at the test track.

4.3.1 Fresh and Hardened Concrete Properties

The Final SCC mix was performed a total of 6 times. The first five were considered trial mixes and were mixed at the Argos Concrete Plant in Jacksonville FL. Table 4.4 shows the test results for slump flow and T-20 for mix 1 and mix 2. The slump flow was consistently retained above 20 inches, and the T-20 was about 4 seconds or less for the duration of 60 minutes. The test results demonstrated the ability of the SCC mix to retain its workability for at least 45 to 60 minutes after the addition of the accelerator providing a continuous and uninterrupted agitation of the mix is maintained. Two possible explanations are offered for the workability retention ability of the SCC mix after the addition of the accelerator. First, the nonstop agitation of the mixes allowed the workability retainer, high-range water reducer, and set retarder admixtures to reach their maximum impact. Second, the continuous agitation may have prevented an early negative impact of the accelerator on the SCC mix setting time and workability. This was further demonstrated by the stable tem-

49 Table 4.4: Slump Flow and T-20 Test Results

Mix 1 Mix 2 Time Min Slump Flow T-20 Slump Flow T-20 (in) (mm) (sec) (in) (mm) (sec) 0* 20 480 9 22 550 1 10 28 710 2 – – – 15 – – – 24 600 2 20 29 737 2 – – – 25 – – – 27 675 3 30 28 700 2 – – – 35 – – – 29 725 3 40 26 648 3 – – – 50 22 546 4 – – – 60 15 380 >10 29 725 3 * This test was performed after 6 minutes from adding accelerator and remixing

perature of the agitated concrete inside the mixer. However, after completing the slump flow test, it was noticed that the remaining mix leftover the wheelbarrow started immediately to lose its workability. This was another indication that the accelerator had an immediate and quick impact on the workability when the SCC mix became stagnant. Also, during the slump flow tests, the SCC mix seemed to maintain good cohesion with no noticeable segregation as evident from Figure 4.5. The absence of any visible segregation may have been the result of using a proper dosage of the workability retaining admixture. This finding was even more interesting when considering the fact that grade 57 aggregate was used in the mix. This size of aggregate is considered by industry to be the upper limit of any coarse aggregate size allowed in SCC mixes before segregation will most likely occur. The ambient temperature during the five trial batches ranged from 75 to 78 degrees Fahrenheit and the concrete temperature in the mixer was about 80 degrees Fahrenheit. All batches included the same accelerator apart from batch 1, where no accelerator was added. Also, due to the small size mixer, it was not possible to prepare sufficient specimens for the workability and for all necessary strength tests. The slump flow/T-20 tests, as well as the 4 and 6 hours compressive strengths tests were performed at the Argos plant laboratory, while the 9 hours tests were performed at the

50 Figure 4.5: Trial Batches at Argos Concrete Plant in Jacksonville, Fl

FAMU-FSU laboratory. Results of the slump flow/T-20 tests and the compressive strengths for the five trial batches are shown in Table 4.5. Despite the variability in the moisture content of aggregates, the slump flow test results ranged from 20 to 30 inches. These measurements were retained for at least 40 minutes after mixing with the accelerator and maintaining agitation of the SCC mix. The concrete temperature during the agitation period did not increase to indicate any significant hydration of the concrete inside the rotated mixer. This was another validation of the FAMU-FSU laboratory test results which showed when SCC was in continuous agitation mode, the concrete mix retained its workability. The compressive strengths after 6 hours for mixes 4 and 5 were greater than 2,200 psi required by the FDOT specification. In Mix 3, the compressive strength was much closer to the required FDOT strength. This was a result of the presence of excessive moist fines in the coarse aggregate. Nevertheless, the Jacksonville batch plant results proved that the SCC mix had the flexibility to accommodate variabilities in commercial concrete plants while still meeting the FDOT strength requirements. It should be noted that the 9-hour specimens in batches 1 and 2 were transported and tested at the FAMU-FSU labs due to the closing of the Argos lab at the end of that workday. The SCC ingredients for the 4.5 cu yd mix were batched with the ambient temperature at the concrete plant was 78o F. After mixing the ingredients, the slump flow was measured at 25 in and

51 Table 4.5: Test Results at Argos Plant Trial Batches

Batch Slump Flow T-20 Comp. Strength No. min in sec hr psi 0 28 4 1* 9 1868 30 21 5.5 0 30.5 4 2 9 4863 10 30 4 20 28.5 3 4 1212 3 30 30 2 6 2459 30 20 4 4 6 3584 40 27 4 0 20 1 5 6 3990 60 14 14 * No accelerator was added to the mix

T-20 at 4 sec. It took about 45 minutes to transport the SCC mix to the test track. At the test track, a second slump flow was performed and measured 19 in. This was a reduction of about 6 in from plant slump measurement. The accelerator was then added to the mixer, as shown in Figure 4.6, followed by 5 minutes of remixing. A third slump flow test was performed and resulted in 21 in. spread in 5 seconds. The truck then discharged the SCC mix to fill the 12ft x 12ft x 8in replacement slab pit. The mix was cohesive and rushed down the mixer chute at a high velocity. No segregation was noticed as the mix moved swiftly and uniformly in all directions to fill the pit. It took 3 minutes to fill the entire test pit. Compared to conventional concrete casting with high early strength mix, the SCC mix showed excellent performance at a much higher discharge rate with no segregation. Once it settled in the pit, the mix started to set quickly, and its temperature begin to elevate rapidly. It was not possible to attain a smooth surface finishing because of the lack of professional finishers at the site (Figure 4.7). However, on job sites with abundance of skilled labor and proper finishing equipment, the slab would have been poured and properly finished in less than 10 minutes. This fast rate of placement will contribute to a much higher productivity compared to conventional, low –slump, high early strength, concrete mixes. It is important to note that once the accelerator was added, the mixer had to continue in nonstop agitation mode, when not discharging concrete, with slightly higher revolution speed than

52 Figure 4.6: Addition of the Accelerator at the Track

Figure 4.7: Surface Finish of Replacement Slab

53 conventional agitation speed, until completely discharging its SCC load. If the mixer was allowed to slow down or stop, the concrete would have set inside the mixer and may require an elaborate and possibly expensive process to dislodge the hardened concrete from inside the mixer drum. Managing the appropriate quantity of the accelerator and withheld mix water in SCC mix for slab replacement was very important to achieve the full advantages of SCC mix use in replacement slab. Based on the research, the amount of accelerator should be reduced when the ambient temperature exceeds 85 degrees Fahrenheit. This may not be an issue cool night paving. However, in daytime paving under hot weather, an excess amount of accelerator could aﬀect mix performance with respect to shorter setting time and rapid loss of workability. The high ambient temperature at the job site will increase in the rate of concrete hydration and the concrete temperature. This can be dealt with by withholding a certain percentage of the accelerator from the mix without aﬀecting the rate of early strength gain. At the test track, six 6in x 12in cylinder specimens were cast to be tested at 4 and 6 hours. Also, the slab was covered with a plastic sheet to maximize curing and to achieve high early strength, as shown in Figure 4.8. The temperature of the concrete was monitored intermittently, using an infrared laser thermometer and a thermal camera. The maximum surface temperature reached about 135degrees Fahrenheit. The six concrete cylinders were transported to Argos laboratory for

Figure 4.8: Slab Curing with Plastic Sheet

54 testing. However, due to the limited access to the laboratory after work hours, only three cylinders were tested after 4 hours. The remaining 3 specimens were transported to FAMU-FSU labs, and were tested after 3 days. The temperature measurements were used to estimate the compressive strength of the SCC at the test track using the maturity relationship of the mix, as will be explained in phase three

4.3.2 Truck Loading of Replacement Slab

Six hours after SCC placement, the replacement slab was loaded using a 25,000 lb truck as shown in Figure 4.9. The truck had single front and double rear axles. The truck represents a typical heavy construction or utility truck on urban streets or county roads. The truck made 100 passes on the slab. During the truck laps, the slab was wetted frequently and checked for crack. No cracks were detected. The 100 passes were completed in about one hour. A ﬁnal close examination of the slab surface did not reveal any damage or cracking in the slab. Following the truck loading,

Figure 4.9: Load Testing of the Replacement Slab Using a 25,000 lb Truck the slab was marked for core sampling at a later time. Nine cores were obtained by a commercial testing lab from various locations including the corners, edges and center of the slab. At the FAMU- FSU laboratory, the cores were examined to determine if any segregation had taken place during casting of the slab. No segregation was detected as evident from the close up in Figure 4.10. In addition, minor bug holes were present on the surfaces of the cores. The samples were later tested at 28 days. The average strength of concrete at 28 days was 10,140 psi.

55 Figure 4.10: Core Samples From the Replacement Slab – No Sign of Aggregate Segregation

4.3.3 Precast Panels Test Results

The objective of measuring the panel displacements was to determine the effectiveness of the proposed panel design in resisting movements under the truck loading, and to establish if impact occurs at the joints between the panels and with the surrounding pavement as a result of horizontal movement of the panels. Eight Linear Variable Differential Transformers (LVDTs) were installed on the panels to measure the vertical and the horizontal displacements, as shown in Figures 4.11 and 4.12. Six LVDTs were positioned to measure vertical displacement panels. The LVDTs were connected to a data acquisition system and a laptop to capture and record the displacements. Two additional readout units were also used to provide real-time monitoring of the vertical and horizontal displacements. These instant readings from the units were helpful to guide the truck during loading and braking actions. Because of the proximity of the LVDTs to the truck wheels, the speed of the truck was maintained below 5 mph to avoid any accidental collision between the truck wheels and the LVDTs. The truck was guided on a specific path to keep it away from the LVDTs. Because of the distance between the front and the rear axles, it was not possible to have the full truckload (60,000 lb.) on one panel at one time.

56 Figure 4.11: Schematic Plan of Instrumentation of the Precast Panels

Figure 4.12: Vertical and Horizontal LVDTs on the Precast Panels

57 4.3.4 Truck Loading on Precast Panels

As mentioned previously, a 60,000 lb. concrete pump truck was utilized to apply both vertical wheel loads and horizontal braking loads on the precast panels, as shown in Figures 4.13 and 4.14. The truck had a single axle under the cabin and tandem double axles in the back. The loading ratio of the front to back axles was 1:3. Accordingly, the vertical load on the front and rear axles were 15,000 lb and 45,000 lb respectively. During displacement monitoring, it was noticed

Figure 4.13: Concrete Pump Truck Loading Precast Panels that after a few loading cycles on the vertical displacements of the panels started to stabilize, an indication of firm seating of the panels in the pit. This was noticeable for measurements from LVDTs CH4-B1, CH2-B1, CH3-B1, CH3-B2, CH2-B2, and CH2-B2 as shown in Figure 4.15. The vertical displacement measurements averaged from 0.11 to 0.30 in. With respect to the horizontal displacement, it was noticed that driving the truck over the panels did not produce significant horizontal displacements (Figure 4.16). The horizontal displacement measurements were induced by the moving truck and averaged from 0.01 to 0.05 in. It should be noted that these displacements did not count for the horizontal displacements due to truck braking. The profile of the vertical displacements is shown in Figure 4.17. These measurements suggest that the permanent settlement “panel seating” under traffic loads upon lane opening would almost cease after a few load repetition from the passing trucks. It is possible that the reduction in the

58 Figure 4.14: Configuration of LVDTs and Wheel Loads settlement rate after a few cycles was due to a combination of the heavy weight of the precast panels (8500 lb/panel) and multiple load application of the truck. Cyclic loading increased the compressibility of the soil, and hence increased the stiffness of the base layer underneath the slabs. The joint faulting was less than 3/16” after the completion of the 50 loading cycles. This faulting is considered acceptable since the function of the panels in the proposed system will be as temporary placeholders prior to casting the permanent slab replacements. Also, it was noticed that there were no permanent vertical or horizontal displacements in the panels after sudden braking of the truck. This demonstrated the efficient design of the precast panels and stability of the leveling course. The four 8 in wide recessed stripes at the bottom of each panel formed ridges on the surface fine recycled aggregate leveling course. These ridges provided interlocking action and frictional resistance at the interface between the panels and the base layer. Upon removing the precast panels, the striped impressions in the leveling course were intact indicating little, if any, horizontal shifting of the panels had occurred, as shown in Figure 4.18. This suggested the fine recycled aggregate layer experienced some densification and re-cementing during the repeated load application. This increased densification led to the reduction in the panel displacements. The vertical and horizontal

59 Figure 4.15: Vertical Displacements of the Two Precast Panels

Figure 4.16: Horizontal Displacement of Precast Panel 2

60 Figure 4.17: Proﬁle of Vertical Displacement of Precast Panel 1

Figure 4.18: Imprints of Ridged Strips on Surface of Recycled Aggregate Leveling Layer

61 displacements due to sudden brake were about 0.18 in and 0.03 in, respectively. The presence of the recessed stripes at the bottom of the panels and backer rod foam around the perimeters of the two panels may have contributed to the low horizontal displacements. It should be noted that the backer rod would most likely act as a cushion in case of any impact between the adjoining panels or at joints with the surrounding pavement. This ﬁeld demonstration validated the robustness of the precast panel design features. The reinforced panels would resist cracking during handling and multiple installations. The recessed stripes at panel bottom surface of the panel would contribute to more interlock and friction that would minimize horizontal movements. The backer rod would provide stability to the panels and preserve their structural integrity during multiple uses, as well as protect the surrounding concrete from damage during panel installation and from traﬃc action.

4.4 Phase Three Results

Phase three will discuss and display the results obtained from the laboratory testing outlined in the prior. The fresh and hardened concrete properties of each mix are also presented in this chapter. The rest of the section focuses on the development of the two maturity curves and how accurate the curves compare to later mixes.

4.4.1 Fresh and Hardened Concrete Properties

Table 4.6: Fresh Concrete Properties for Argos Mix

Argos Mix Size Ambient Temp Concrete Temp Slump ft3 °F °F in. 1 3 92 101 16.5 2 3 108.6 97.9 20 3 3 83.4 85.5 18 4 3 89 93 22 5 3 89 93.9 20

The fresh concrete properties are considered the properties on the concrete before it is set in place. This includes mix duration, slump results, ambient temperature, and laboratory temperatures. Fresh concrete properties were taken to make sure that the mixes are repeatable for later

62 Table 4.7: Hardened Concrete Properties Argos Practice Mix 1

Argos Practice Mix 1 Age Humboldt FLUKE FLIR IR Gun Strength Temp Temp Temp Temp hr °F °F °F °F psi 0 102.2 89.0 89.0 89.0 – 3 101.3 92.6 92.0 94.8 – 4 101.3 109.6 109.7 109.7 1598 5 98.6 111.0 112.1 112.8 2555 6 95.9 109.1 109.7 110.8 3387 7 93.2 107.4 106.7 108.5 4033 8 91.4 105.2 105.4 106.3 4086 9 89.6 102.2 102.7 103.1 4219

experiments and met the ASTM standards mentioned in previous chapters. All the fresh concrete properties can be found in Table 4.6. The hardened concrete properties are the properties of the concrete after it has set and started to cure. The maturity curves developed are based on these properties. The properties discussed in the following tables include compressive strength and temperature readings. As mentioned before the concrete strength was tested every hour after it was determined strong enough to test (3 hours after pour). The temperature was taken automatically and manually every hour starting when the concrete was set in place. The results can be seen in Table 4.7. The two mixes that developed the maturity cures did not use the thermal imaging or infrared devices. The mixes instead follow the traditional maturity method found in ASTM C1074-11. The compression strengths of each mix can be found in Table 4.8.

4.4.2 Development of Maturity Curves

For this section, two maturity curves had to be developed and tested for two diﬀerent but similar concrete mixes. To accomplish this, several mixes were completed using the guidelines described in ASTM C1074-11. A novel approach was also taken using several other temperature reading devices. To develop the maturity curves, there are three main ingredients. The ﬁrst ingredient is time; the time must be noted when the concrete has been set in place, and every time temperature is taken. For this ingredient, a maturity meter was used, and it logged temperature data every 30 minutes.

63 Table 4.8: Hardened Concrete Properties Argos Mix 2 and Mix 3

Argos Practice Mix 2 Argos Practice Mix 3 Age Humboldt Strength Humboldt Strength Temp Temp hr °F psi °F psi 0 92.3 – 89.6 – 2 94.6 – 96.8 – 3 104.9 495 115.7 1010 4 112.1 1466 122.0 1656 5 116.6 2481 125.3 2626 6 119.3 3488 125.6 3223 7 119.3 4151 124.7 3684 8 119.3 5086 122.9 3977 9 118.4 5365 122.0 4062 10 115.7 5524 118.4 4119 11 113.9 5439 115.7 4255 12 112.1 5644 110.3 4820 24 92.3 6622 82.4 5462 3 day 77.9 8212 78.8 5931 7 day 77.9 7869 77.0 6435 14 day 77.0 9160 78.8 6797 28 day 78.8 10538 78.8 6931

The second ingredient is temperature; temperature of the concrete must also be recorded. This is usually recorded in a specified interval. For this study, every hour was chosen this is because the concrete being used is fast curing and changes temperature rapidly. The last ingredient needed to create the maturity curves is compressive strength. Strength of the concrete is compared to maturity (TTF). Graphically, maturity is created with the first two ingredients. Temperature from the thermal couples were collected every hour. This makes up the first two columns in the Table 4.9. The third column is an average temperature over the specified period of time. Therefore the temperature reading from one reading is averaged with the following reading one hour later. The fourth column in the maturity function is used to develop maturity of the specified time interval. The last column is the cumulative amount of maturity over time. This method was applied for each mix and developed maturity curves shown in the following figures. The maturity curve above in Figure 4.19 shows the relationship between strength and TTF. The equation attached has a close relationship and can be used to predict the strength of the in-place

64 Table 4.9: Example Maturity Data from A-Materials

Example Maturity Data from A-Materials Time Thermocouples Average Temp (Avg Temp -14) ∆ t Maturity Strength °F hr °F °F TTF psi 0 89.6 89.6 – 0 – 1 91.4 90.5 76.5 76.5 – 2 96.8 94.1 80.1 156.6 – 3 115.7 106.25 92.25 248.85 1010.3 4 122.0 118.85 104.85 353.7 1655.5 5 125.6 123.80 109.8 463.5 2626.5 6 125.6 125.6 111.6 575.1 3223.4

Figure 4.19: Argos 28 Day Maturity Curve

65 concrete, however it will underestimate the strength. To make the comparison more visible for early strength another graph was created to show a closer look into the ﬁrst 12 hours as shown below in Figure 4.20. Similarly two graph were created for the A-Materials mix including a graph that shows the 28 day maturity curve (Figure 4.21) a close up of the ﬁrst 12 hours (Figure 4.22).

Figure 4.20: Argos Maturity Curve (First 12 Hours)

With the maturity curves developed, the next step was to test the accuracy of them by repeating the two mixes and testing different parts of the curves. As mentioned before, different desired strengths were selected to confirm or deny the curve. The curve could not exceed 10% difference from the predicted strength versus actual strength. If the curve did exceed the 10% limit a new maturity curve would need to be created (ASTM Standard C1074-11, 2011). Two mixes were done 10 minutes apart to test the two maturity curves the first mix. The maturity curves were then printed out and marked so that the desired strength matched up to the predicted maturity values as shown in Table 4.11 . Figures 4.23 and 4.24 display how the curves were marked to get the theoretical maturity value for the desired strength. To test the theoretical maturity values for accuracy, 44 cylinders where casted for each mix. Two cylinders from each mix had thermal couples impeded into them and connected to a maturity

66 Figure 4.21: 28 Day A-Materials Maturity Curve

Figure 4.22: 12 Hour A-Materials Maturity Curve

67 Figure 4.23: Argos Theoretical Maturity Values meter. This meter had the ability to display maturity values for 30 minute intervals. When the actual maturity was close to the theoretical maturity value, five cylinders where tested in a row that covered about a ten minute span of time. Testing this many cylinders gave enough elapsed time to hit the target maturity. For example, on the first desired strength of 1600 psi, the theoretical maturity was 347 for A-Materials. Since the maturity meter only takes a reading every 30 minutes at four hours, the reading was 337.05 and at 4 and a half hours it was at 390.38. Thus the desired maturity value was some time in between the two readings. This is the reason why five cylinders were tested in a row. The results from the accuracy testing can be found in Table 4.11. Mix 4 and Mix 5 were performed to test the accuracy of the two maturity curves developed based on the concrete properties mentioned in Table 4.11 were not tested every hour like the previous mixes, but instead tested at different maturity readings that predicted a desired strength. The crucial strength of this type of concrete usage is 2200 psi within 6 hours, therefore three points were taken around 2200 psi mark (1600 psi, 2200 psi, 3500 psi). To test the later strength of the concrete 5800 psi was also completed. As shown in Table 4.11, the percent errors for Argos are within the acceptable range established by ASTM C1074-11. However, the A-Materials mix had

68 Figure 4.24: A-Materials Theoretical Maturity Values

Table 4.10: Final SCC Mix Design Argos Versus A-Materials

69 Table 4.11: Maturity Predictions for Argos Mix 4 and A-Materials Mix 5

Desired Strength Predicted Maturity Actual Strength Percent Error Argos A-Materials Argos A-Materials Argos A-Materials psi TTF TTF psi psi % % 1600 340 347 1437 1863 10.2 16.4 2200 414 414 2440 2900 10.9 31.8 3500 540 645 3790 4258 8.3 21.7 6400 2200 2200 6755 6436 5.6 0.6

much higher percent error, and this indicates that the maturity curve needs to be recreated. The A-Materials maturity curve underestimates the concrete strength. This could been a result of the concrete getting hotter than the previous mixes. As mentioned before, several other devices where used to record temperature data. The information collected was compared to the customary method of using a maturity meter. The data collected was from the first mix and mix number 4. Both of these mixes were Argos so that maturity curves can be compared to see how different the curves can be just by changing the device. The first device compared to the maturity meter was the IR gun, this device does not automatically record the temperature, and it can only read the surface temperature. Therefore, temperature readings were manually taken every hour and recorded into a spreadsheet were a maturity curve was developed. Two other devices chosen were thermal cameras one being a high-end camera that takes high resolution images, and the other being an iPhone attachment. The data gathered concentrated on the first 6 hours of curing seen in Table 4.12. As seen in the Figure 4.25, the different sensors have a strong correlation to the Maturity meter. The trend line is also close in relation. It is noted that the maturity meter seems to predict the strength earlier than the other devices, this is probable because the maturity meter records the internal temperature while the others record the surface temperature. For mix number 4, maturity was collected every hour using the same method as the first mix. Table 4.12 compares the maturity value from each mix. With the limited amount of strength data from the mix 4, the maturity curve can only be represented by 3 points in the first 6 hour, however it can be seen in the following chart the devices share a common result similar to one found in the first mix Figure 4.8. The Maturity meter indicates a larger TTF value for the

70 Table 4.12: Novel Approach Maturity Comparison

Device Maturity Mix 1 Argos (TTF) Hour 0 1 2 3 4 5 6 IR Gun 0 NA NA 233.7 322.0 419.2 517.0 FLIR Thermal Camera 0 NA NA 229.5 316.4 413.2 510.1 FLUKE Thermal Camera 0 NA NA 230.4 317.5 413.8 509.85 Maturity Meter 0 263.3 350.6 436.5 519.75 Maturity Mix 4 Argos (TTF) Hour 0 1 2 3 4 5 6 IR Gun 0 77.8 160.2 248.9 345.4 441.2 536.2 FLIR Thermal Camera 0 76.7 159.1 247.7 341.4 431.9 527.7 FLUKE Thermal Camera 0 77.8 164.8 258.3 359.4 458.6 557.1 Maturity Meter 0 83.25 169.2 262.1 360.2 463.5 565.4

Figure 4.25: Maturity Relationship from Diﬀerent Sensors Mix 1

71 Figure 4.26: Maturity Relationship from Diﬀerent Sensors Mix 4 same compressive strength value. This is perhaps why the maturity meter is the recommend device to use when developing maturity curves.

72 CHAPTER 5

DISCUSSION AND CONCLUSION

5.1 Project Summary

Slab replacement is a major activity in any concrete pavement rehabilitation project. However the current process is slow for such a small concrete project which ranges between 25 to 50 cu. yds. The low number of slab replacements completed during the road closure period is due to the short window available to complete the multiple, time-consuming activities required for each slab replacement. To increase productivity, the concrete mix is often designed to achieve the lane opening strength requirement of 2,200 psi in the shortest possible time to enable placing more slabs during a lane closure period. This requires the use of excessive quantities of cement and/or accelerators to achieve the required strength. A likely consequence of such action has been the development of premature cracking in many newly replaced slabs. This project was initiated with the objective to develop a method using temporary reusable precast panels and self-consolidating concrete (SCC) to accelerate the slab replacement process. The goal of the method is to increase contractors’ productivity, reduce construction completion time, reduce project cost, and minimize premature cracking. The three components of this method, namely, reinforced precast panels, SCC mix and nondestructive strength testing were designed, prepared, and tested in the lab, and then demonstrated their viability in the ﬁeld. Construction criteria, detailed guidelines and speciﬁcations were also developed to facilitate implementation of this method in concrete pavement rehabilitation projects. The method involves installing one or two precast panels in the replacement slab pit after removing the deteriorated section of the original slab. The panels are kept as place holders for a few days and then removed. The open pit is then cast using SCC mix to form a permanent replacement slab. The method was developed and tested and is summarized in the following steps:

1. Remove the deteriorated section of the pavement slab.

2. Drill holes for the new dowel bars in the appropriate sides of the replacement pit, then clean and patch the holes with tape.

73 3. If the depth of the resulting pit does not match the thickness of the precast panels, add a leveling course to the bottom of the pit or, otherwise, excavate the bottom.

4. Install one or two precast panels in each slab replacement pit.

5. Remove the precast panel(s) and transport to the next segment of the project for re-installation.

6. Remove the tape that covered the dowel holes and install new dowel bars in the epoxied holes.

7. Cast the replacement slab using the SCC mix.

8. Level, ﬁnish and cure the slab surface.

9. Repeat the construction sequence (steps 1-9) until completing the required slab replacements for the project.

5.2 Conclusions

The following sub sections are conclusions of each major component of this project.

5.2.1 Self-Cosolidating Concrete (SCC) Mix

The following list describe individualized conclusions that are based on the results and analysis of the SCC mixes performed in the laboratory and in the ﬁeld.

1. An innovative SCC mix was developed for slab replacements in concrete pavements. The high flow rate of the SCC mix for rapid discharge and finish without segregation will likely increase productivity of contractors during lane-closure time and will reduce maintenance of traffic (MOT) time.

2. The SCC mix using Type I/II cement, 57 grade aggregates, silica sand, low w/c as well as a combination of high range water reducer, workability retainer and accelerator produced a highly workability mix for rapid discharge and shorter casting time, and exceeded the 2,200 psi strength required by the Florida DOT for lane opening. This SCC mix is a departure from conventional SCC mixes used in structural applications.

3. The SCC mix design was highly eﬀective for casting a slab replacement in the ﬁeld without the need for compaction. This will potentially save time and manpower.

4. The concrete workability was retained for at least 60 miutes after adding and mixing the accelerator. The workability retention will allow the concrete in truck mixers to maintian a high workability rate during truck idle times between successive concrete placments and achieve rapid discharge during casting of isloated replacment slabs.

74 5. Vibrated and nonvibrated samples cast from the same SCC mix showed very small diﬀerences in compressive strength and weight. As a result, no concrete vaibration was needed during slab placments or in preparing cylinder samples.

6. There was no aggregate segregetion in any of the batched mixes despite high workability and the use of grade 57 aggregate in the mix.

7. There was no segregation present in the 4 in concrete cores taken from the 12 in x 12 in x 8 in replacment slabs.

8. It took less than three minutes to completely cast a 12’ x 12’ x 8” replacment slab.

9. Adjustments to the admixture dosage rates may be necessary when the SCC mix is speciﬁed for longer concrete transportation periods, when the admixture source changes.

10. Adjustments in the mix cement content and/or accelerator dosage rate may be necessary when ambient temperature is higher than 85oF or lower the 50 oF to maintain high workability and agency-required strength for lane-opening.

11. Strength of the SCC mix at 6 hours greatly exceeded the required 2,200 psi strength for lane opening.

12. It is very importance to maintain the truck mixer in agitation mode (when not discharging concrete), non-stop, and at a slightly higher speed than conventional agitation. This will prevent premature setting of the SCC mix after addition of the accelerator at the jobsite.

13. Immediate leveling and ﬁnishing of the cast slab will be required due to rapid loss of SCC workability upon discharge in the replacement slab pit.

5.2.2 Temporary Reusable Precast Panels

The following list presents the conclusions made during the testing and installation of the temporary and reusable precast panels.

1. Two 6’x 12’x 8” panels were designed using double mats of no. 4 reinforcing bars. Both panels showed robust performance in laboratory and ﬁeld demonstrations.

2. The imbedded backer rod gaskets along the sides of the panels facilitated installation in and removal of panels from the replacement slab pits. Lubricating the gaskets expedited the installation process. The gasket also would soften any impact of the panels against the surrounding concrete and prevent damage to both sides.

3. Maintaining a 1.5” gap between the panels and the surrounding concrete made installation simple and less likely to damage surroundings.

75 4. The recessed stripes along the bottom surface of the panels provided stability and strong interlock with the base to resist panel shifting under traﬃc.

5. The panels displaced 3/16 in. vertically and had very low horizontal displacement when subjected to 50 load repetitions from a passing 60,000 lb truck. This faulting is considered acceptable since the panels are temporary place holders and would remain the pit for only a maximum of a few days.

6. The installation time for each panel was approximately 10 minutes and its removal was nearly 5 minutes. With more dedicated lifting and maneuvering equipment, these times can be shortened further.

5.2.3 Nondestructive Strength Testing (Maturity Curves)

After testing the SCC mixes from Argos and A-Materials against the maturity curves, the following conclusions were made.

1. Two maturity curves were developed and tested using ASTM Speciﬁcation C1074-11.

2. Two diﬀerent mix designs were tested, using material from two locations Argos and Anderson Materials (A-Materials).

3. The Maturity curves were checked for accuracy, with the Argos mix design proving to be within the 10 percent of actual strength values.

4. A novel approach of maturity curves showed that with additional testing they can be proven accurate.

5. With each adjustment to the mix design a new maturity curve will have to be developed.

6. Maturity curves are most accurate using imbedded thermal couples.

5.3 Recommendations for Further Study 5.3.1 General

This slab replacement method is most cost eﬀective and eﬃcient when:

1. The volume of concrete speciﬁed in the contract bid is more than 1000 cu. yds.

2. A limited window for lane closure is speciﬁed, especially during nighttime.

3. Signiﬁcant premature cracking is observed using conventional concrete mixes with excessive amount of cement and accelerator to achieve required strength in less than 4 hours from the lane opening time.

76 4. Many deteriorated slabs that require removal and replacement are located in close proximity of each other.

5. The cost beneﬁt analysis favors the use of this method.

Other favorable factors include a close distance of a concrete batch plant and contractor’s storage yard to the project site, and safety concerns of having too many workers at slab replacement sites.

5.3.2 Self-Consolidating Concrete (SCC) Mix

Based on the research and the conclusions made in this report, the following practices regarding the use of SCC mixes for concrete pavement slab replacement are recommended.

1. Modify FDOT speciﬁcation 353 to allow the use of SCC mix in slab replacements. Allow the use of 67 or 57 grade aggregates and accelerator admixture in the SCC mix.

2. Use the SCC mix design developed in this project or an nequivalent mix that prodces a minimum 20” slump flow immediately after addition of the accelerator. Make adjustemts in specification 353 to reflect this property.

3. The SCC mix must maintian a high slump ﬂow for quick dischrge and no vibration, and develop a minimum strength of 2,200 psi at lane opening.

4. There will not be a need for internal vibration during casting of the replacment slab using the highly workable SCC mix.

5. Use manual distribution of the concrete and apply internal vibration if the ﬂow of the SCC mix appears to have decreased signiﬁcantly.

6. Consider starting the roller or surface vibratory screed to achieve a smooth surface finish while the concrete is being discharged in the replacment slab pit or as soon as the pit is filled. Delaying the surface finish may present challenges to proper surface finish as the mix will harden rapidly as the mix is settled in place.

7. Upon addition of the accelerator to the truck mixer, maintian the mixer at a continous and no-stop agitation mode (when not discharging concrete). Increase the revolution speed of the mixer to ensure that the high workability of the mix is mainitined, until its entire concrete load has been discharged.

8. When the full slab is removed, the SCC mix may be used to cast a new slab without the need to include the precast panel component of the method.

77 5.3.3 Temporary Reusable Precast Panels

In order to use temporary reusable precast panels in concrete pavement slab replacement projects, this research recommends to:

1. Revise FDOT Standard Index 308 for 20 ft long slabs to require the designer to use only four standard dimensions for the deteriorated area to be removed. These dimensions shall be, 6’ x 12’, 8’ x 12’, 10’ x 12’ and 12’ x 12’. Cracks that exceed 12’ shall require replacement of the entire slab. For 15’ long slabs, use three standard dimensions, 6’ x 12’, and 8’ x 12’ and 10’ x 12’. Cracks exceeding 10’ would require replacing the entire slab.

2. Revise FDOT speciﬁcation 353 to allow the use of precast panels approved by the Engineers as temporary ﬁllers of slab replacement pits.

3. Prepare reusable precast panels in two dimension, 6’ x 12’ x T and 4’ x 12’ x T. Determine the thickness (T) of the panels from the average thickness of representative pavement cores. These panels would ﬁt individually or in a set of two in the proposed standard pits recommended for Index 308.

4. Use the panel design developed in this project or a similar design that produces panels that are robust during handling and transportation, stable under traﬃc, and structurally sound for multiple uses.

5. The precast panels may be reinforced and cast at the contractor’s yard or fabricated in a precast plant and delivered to the contractor. The precast panels may be pretensioned and can be made using lightweight concrete (after testing).

6. Maintain a 1.5” gap between the panels and surrounding concrete to facilitate installation and removal of the panels. This can be achieved by increasing the dimensions of the replacement slab pit of shortening the dimensions of the panels.

7. When the depth of a replacement slab pit is diﬀerent than the thickness of the precast panels by more than 0.25”, add a leveling course material or remove base material (whichever condition exists).

8. Use an eﬃcient equipment to handle the panels at the jobsite, including, loading and unload- ing, as well as installing and removing. Equipment options include, a crane, excavator bucket or a tractor with a telescopic extension boom that can be adjusted in vertically or horizontal reach, or a similar machine capable to lifting up to a 10 ton weight.

9. Replace a precast panel when upon multiple uses becomes severely damaged. A severe damage includes, a crack wider than 0.25” or multiple cracks developed in the panel, or when a spall larger than 4” x 4” is formed in the panel.

78 5.3.4 Nondestructive Strength Testing (Maturity Curves)

When using maturity curves to predict the strength of the SCC concrete replacement slabs, this research recommends to:

1. Revise FDOT standard index 353 to allow the use of the Maturity method

2. Test concrete samples before opening lanes to traﬃc when weather conditions change rapidly over the concrete placement window.

5.4 Recommendations for Future Research

The following issues should be considered for further research:

1. In this study, only two sets of SCC admixtures were used. It is recommended that other concrete admixtures from other vendors be investigated to come up with speciﬁed SCC mixes regardless of the vendors.

2. Precast slabs can be developed from lighter and more durable materials other than concrete. For example, the use of recycled plastic or timber panels can be considered for future research.

3. The current concrete maturity method was found to be limited to the locations of embedded thermocouples. However, the thermal imaging method used in this study showed that for concrete slabs the temperature distribution in the concrete would vary from point to point. Therefore, the thermal imaging technique developed in this study should be used along with the embedded thermocouples to come up with a more rigorous method for concrete maturity measurement.

79 BIBLIOGRAPHY

[1] Barnhart V. Buch, N. and R. Kowli. Precast concrete slabs as full-depth repairs-michigan experience. Technical report, Transportation Research Record 1823, TRB, 2003.

[2] C. Camp. Hydration of portland cement. Technical report, University of Memphis., 2015.

[3] N. Carino. The maturity method: From theory to application. Technical report, National Insitute of Standards and Technology, 2001.

[4] ASTM International. Astm standard c1074-11. standard practice for estimating concrete strenght by the maturity method. Technical report, ASTM International, 2011.

[5] J. Armaghani K. Tawﬁq. Accelerated slab replacement using temporary precast panels and self-consolidating concrete. Final Report BDV30-977-02, FDOT, 2016.

[6] D Khayat, K. Mitchell. Self-consolidating concrete for precast, prestressed elements. Technical report, National Cooperative Highway research Program, 2009.

[7] S. Kosmatka and W. Panarese. Design and control of concrete mixtures. Technical report, Portland Cement Association, 2002.

[8] McCullough F. Burns N. Merritt, D. Precast prestressed concrete pavement pilot project near georgetown, texas. Technical report, Transportation Research Record 1823, TRB, 2003.

[9] Fort Miller. Super slab, 2015.

[10] NRMCA. Curing in-place concrete. Technical report, National Ready Mixed Concrete Asso- ciation, 2000.

[11] A. Nykanen. Hardening of concrete at diﬀerent temperatures, especially below the freezing point. Technical report, Danish Institute for Building Research, 1956.

[12] PCA. The link between concrete sustainability and curing. Technical report, Portland Cement Association, 2015.

[13] J. Plowman. Maturity and the strength of concrete. Technical report, Magazine Concrete Research, 1956.

[14] David Darwin Sidney Mindess, J. Francis Young. Concrete. Technical report, University of British Columbia ,University of Illinois, University of Kansas, October 2003.

80 [15] Ye D. Tayabji, S. and N. Buch. Precast concrete pavement technology, shrp 2 report s2-r05- rr-1. Technical report, Transportation Research Board, 2013.

[16] Shah S. Grove J. Wang, K. Self-consolidating concrete application for slip-form paving: Phase ii. Technical Report , Report No. DTFH61-06-H-00011, National Concrete Pavement Technology Center for Iowa State University, May 2011.

[17] F Yang. Self-consolidating concrete, ce 241. Technical report, University of Berkley California, 2004.

81 BIOGRAPHICAL SKETCH

Steven Squillacote, Jr. was born in Ormond Beach, Florida to Steve and Betsy Squillacote. He grew up working for his father in his general contracting business and knew from a young age that he wanted to be a Civil Engineer. Steven graduated from the FAMU-FSU College of Engineering with his Bachelors in Civil Engineering in December 2013. While working for the FAMU-FSU College of Engineering, he started working for his Master of Science in 2014. In his spare time, Steven likes to help the students in the American Society of Civil Engineers build a concrete canoe in which he hopes that they place ﬁrst every year. Steven also likes to cook, build, create, and enjoy life with friends and family.