Citation
Mechanical and bonding properties of polyester polymer concrete bridge deck overlay

Material Information

Title:
Mechanical and bonding properties of polyester polymer concrete bridge deck overlay
Creator:
Alethafa, Marwa Foad Manher
Place of Publication:
Denver, Colo.
Publisher:
University of Colorado Denver
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Master's ( Master of Science)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Civil Engineering, CU Denver
Degree Disciplines:
Civil enginering

Record Information

Source Institution:
Auraria Library
Holding Location:
Auraria Library
Rights Management:
Copyright MARWA FOAD MANHER ALETHAFA. Permission granted to University of Colorado Denver to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item has the following downloads:


Full Text
MECHANICAL AND BONDING PROPERTIES OF POLYESTER
POLYMER CONCRETE BRIDGE DECK OVERLAY
by
MARWA FOAD MANHER ALETHAFA B.S., University of Al-Qadisiya, Iraq, 2010
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Civil Engineering Program
2016


This thesis for the Master of Science degree by Marwa Foad Manher Alethafa has been approved for the Civil Engineering Program by
Frederick Rutz, Chair Chengyu Li Nien-Yin Chang
Date: December 17, 2016


Alethafa, Marwa Foad Manher (M.S., Civil Engineering)
Mechanical and Bonding Properties of Polyester Polymer Concrete Bridge Deck Overlay Thesis directed by Associate Professor Chengyu Li
ABSTRACT
Polymer Concrete (PC) is a composite material made from a combination of mineral aggregate such as sand and gravel with a polymerizing monomer. First, it has been used in many different areas such as patching, overlays for highway pavements and bridge decks, and flooring. In recent years, polyester polymer concrete has become widely used in bridge deck overlay. This article will investigate the material properties of the composite section with a variety of tests including cylinder compression, beams bond strength, shear tests, and fatigue tests. Additionally, the results obtained from those tests are critical to know the mechanisms of polyester polymer concrete.
The form and content of this abstract are approved. I recommend its publication.
Approved: Chengyu Li


ACKNOWLEDGEMENTS
There are numerous people that in one way or another have helped me make this research successful. First of all, I would like to thank Dr. Chengyu Li who has served as an advisor for this research. His enthusiasm and dedication have been really motivating, and he has been very supportive in every stage of the research, from planning and preparation to testing and writing this thesis.
I would like to thank Tom Thuis, Peter Sillstrop and Jac Corless at the University of Colorado Denver Electronic Calibration and Repair Lab for being so generous with me and training me on MTS machines. I would also like to express my gratitude to Sheila Cherry who is a regional manager at Kwik Bond Polymers. Sheila has done a lot supporting and pushing this thesis forward; her presence to apply the overlay, providing all mixture materials, and offering all the equipment for the pull-off test. At last my husband and fellow graduate student, Hussein Jaaz, who spent several days helping me with casting concrete, coring pull-off samples, and saw cutting the shear test pads.
IV


TABLE OF CONTENTS
CHAPTER
I INTRODUCTION..........................................................1
1.1 Overview..........................................................1
1.2 Research Obj ectives..............................................2
II LITERATURE REVIEW....................................................3
2.1 Introduction......................................................3
2.2 Bridge Deck Types.................................................5
2.2.1 Polymer concrete overlay with concrete deck...................5
2.2.2 Polymer concrete overlay with steel deck.....................7
2.3 Bond Strength....................................................10
2.3.1 Pull-off Test Failure modes..................................13
III EXPERIMENTAL WORK....................................................15
3.1 Methods..........................................................15
3.2 Samples Preparation..............................................15
3.2.1 Concrete samples.............................................15
3.2.2 Concrete curing..............................................19
3.2.3 Steel plates preparation.....................................20
3.2.4 Surface preparing............................................20
3.2.5 Applying polymer concrete layer..............................22
3.3 The Machines That Were Used......................................24
IV TESTING..............................................................26
4.1 Pull-off Test....................................................26
v


4.2 Compression Test................................................31
4.3 Bending Test....................................................41
4.4 Shear Test......................................................47
4.5 Fatigue Test....................................................54
V DISCUSSION..........................................................60
5.1 Pull-off Test Results...........................................60
5.2 Compression test Result.........................................63
5.3 Bending Test Results............................................63
5.4 Shear Test Results..............................................64
5.5 Fatigue Test Results............................................64
5.6 Final Conclusion................................................65
5.7 Farther Research................................................66
REFERENCES..............................................................67
APPENDIX
A. Product Data Sheet: PPCTM 1121 ..................................68
vi


LIST OF TABLES
TABLE
4.1: Results of 18 pull off tests have been done for a verity of samples.......28
4.2: Results of compression tests..............................................37
4.3: Results of all four bending tests.........................................47
4.4: Concrete cylinders results...............................................49
4.5: Direct shear test results.................................................50
4.6: Fatigue tests events summary..............................................56
5.1: Comparing the tensile and compressive strengths for concrete and polymer.62
vii


LIST OF FIGURES
FIGURE
2.1. Conventional concrete bridge deck............................................4
2.2. Polyester polymer concrete overlay...........................................4
2.3. Applying polyester polymer concrete overlay on Newburgh-Beacon Bridge new
concrete deck in Beacon, NY................................................6
2.4. Repairing and later applying overlay for Carlson bridge existing concrete deck in
Illinois Route 7 over the Des Plaines River in Lockport, Illinois..........7
2.5. Preparing Brooklyn Bridge orthotropic steel deck in the shop.................8
2.6. New Brooklyn bridge orthotropic steel deck...................................8
2.7. Bonding stress distribution.................................................10
2.8. Diagram of overlay and substrate............................................11
2.9. Concrete Pull-Off Testing Schematic.........................................13
2.10. Failure modes of the pull-off test..........................................14
3.1. Molds are ready before the casting day......................................16
3.2. The ready mixed concrete that was used in samples casting...................17
3.3. Slump test..................................................................17
3.4. Compacting the first layer of concrete panel................................18
3.5. One of the 6 concrete slabs that have been casted...........................18
3.6. First day of concrete curing................................................19
3.7. Curing process at 7 days old................................................19
3.8. Steel plate a day before applying the polymer overlay.......................20
3.9. The panel's surface after it has been roughed up............................21
3.10. Appling zinc primer coat on the steel plate surface.........................22
viii


3.11. Polyester mixture preparing.......................................................23
3.12. Polymer applying process..........................................................24
3.13. DYNA Z Pull-Off Tester............................................................25
3.14. MTS 810 load frame................................................................25
4.2. Attaching the test disc to DYNA Z Pull-Off tester.................................27
4.1. A test disc glued to a concrete specimen..........................................27
4.3. The 10 concrete pull off samples..................................................29
4.4. The 8 steel pull off samples......................................................30
4.5. Stress-strain curve for cylinder P 1-1 @7 days old................................32
4.6. Stress-strain curve for cylinder P 2-1 @ 7 days old...............................32
4.7. Stress-strain curve for cylinder P 4-1 @ 7 days old...............................33
4.8. Stress-strain curve for cylinder P 1-2 @ 24 days old.............................33
4.9. Stress-strain curve for cylinder P 2-2 @ 24 days old.............................34
4.10. Stress-strain curve for cylinder P 3-1 @24 days old...............................34
4.11. Stress-strain curve for cylinder CDOT-1 @ 158 days old.......................35
4.12. Stress-strain curve for cylinder CDOT-2 @158 days old.......................35
4.13. Stress-strain curve for cylinder CDOT-3 @ 158 days old.......................36
4.14. Stress-strain curve for cylinder CDOT-4 @158 days old.......................36
4.15. Preparing polymer samples for compression test and checking the level of their
surfaces by passing them on a concrete saw to grind any excess materials..........38
4.16. The laser extensometer measures the deformation of the cylinder...................38
4.17. Sample Pl-1, the 7 days old, under the load and its failure shape.................39
4.18. Sample P3-1, the 24 days old, under the load and its failure shape................39
4.19. Sample CDOT 1, the 158 days old, under the load and its failure shape.............40
IX


4.20. Sample CDOT 3, the 158 days old, under the load and its failure shape.....40
4.21. Four points flexure test setup............................................41
4.22. Modulus of Rupture-Deflection graph of first beam sample...................43
4.23. Modulus of Rupture-Deflection graph of second beam sample..................43
4.24. Modulus of Rupture-Deflection graph of third beam sample...................44
4.25. Modulus of Rupture-Deflection graph of fourth beam sample..................44
4.26. Failure mode of beam sample #1...........................................45
4.27. Failure mode of beam sample #2...........................................45
4.28. Failure mode of beam sample #3...........................................46
4.29. Failure mode of beam sample #4...........................................46
4.30. Preparing shear samples and the framework for the test....................48
4.31. Sample and frame set-up in the 220 kips MTS machine.......................50
4.32. Sample Sl-1 when the determined area got separated by direct shear........51
4.33. Sample S 5-5 when the determined area got separated by direct shear.......51
4.34. All shear samples after all direct shear tests have done...................52
4.35. Concrete cylinder sample C 3-1 during test it for compressive strength.....52
4.36. Concrete cylinder sample C 6-1 during test it for compressive strength.....53
4.37. Six concrete cylinder sample failure shape under compression test..........53
4.38. Attaching a framework to the 20-kip machine...............................55
4.39. Both samples during their tests...........................................55
4.40. Graph for the first fatigue test shows the relation between applied load and the
number of cycles.........................................................58
4.41. Graph for the second fatigue test demonstrates the relation between applied load
and the number of cycles.................................................58
x


4.42. Both samples after they have been tested.
59
xi


CHAPTER I
INTRODUCTION
1.1 Overview
Polyester Polymer Concrete (PPC) has been used as an overlay for bridge decks since 1970s in the United States. In the beginning, it had been used for rehabilitation of damaged concrete decks, but today, its main use is in bridge overlays. Moreover, because of its many benefits, PPC usage expanded to many other industries such as precast, manufacturing panels, underground vaults and utility boxes, and even railroad ties.
Broadly, polymer concrete overlay replaces the traditional asphalt overlay in highway bridges. One competitive advantage of PPC is that polymer concrete is lighter than concrete and can be applied to a thinner cross sectional area which results in less dead load on the structure. This thin layer of PPC protects deck steel reinforcement from traffic abrasion and environment impacts effectively.
The use of resin in PPC mix makes the polymer concrete a waterproof, durable and long lasting surface. We wanted this overlay because it is less permeable than asphalt and water is the enemy of a concrete deck, said Joseph Ruggiero, the New York State Bridge Authoritys executive director, talking about Newburgh-Beacon Bridge which opened to traffic in 2010 with polymer concrete overlay.
With all the benefits of PPC, there are some disadvantages. The main drawback is its high cost compared to conventional options. A cost versus service life study must be considered to decide if the polymer concrete is the right choose to go with. Another issue when using polymer concrete overlay is the weather at the time of application, where
1


moisture is a significant construction and quality issue; it needs to be monitored before, during, and after applying the overlay to ensure an effective bond.
1.2 Research Objectives
The mechanical properties of polymer concrete are playing a significant role in designing its structural applications. Thus, the aim of this paper is to study the mechanical properties of polymer concrete when it is used as an overlay for bridge decks.
The strategy of this research is to do multiple tests for the polyester polymer in order to study compressive strength, flexural strength, bonding strength, shear strength, and fatigue response. This study also investigates the effect of using zinc primer on steel plate decks with polymer overlay.
Toward this end, ten different panels of the polyester polymer concrete overlay with conventional concrete and steel plates have been prepared to see the mechanisms of polymer concrete. The tests are conducted on six concrete pads and four steel plates two with Zinc and two without Zinc. Cylinder compression, beam tests and bond strength, shear tests, and fatigue tests are compared and analyzed.
The experimental data include both the effect of using zinc primer and high strength concrete deck. Therefore, these data are useful in the evaluation of the mechanical properties of the polymer concrete and application as bridge deck overlay.
2


CHAPTER II
LITERATURE REVIEW
2.1 Introduction
The polymer concrete usage is growing rapidly in many structural and construction applications including bridges. Recent studies mentioned the high effectivity of using polymer concrete in precast bridge components such as deck panels.
For many years in the United States, polymer concrete has been very effectively used in the repair and overlay of damaged cement concrete bridge decks as an adamant and durable material.
One of the biggest advantages of using polymer concrete in many construction applications is the fast curing time of this product which cures in a few minutes or hours while concrete components cure in a few days or weeks. Add to that this material produces lightweight structures compare with asphalt and concrete materials due to using thinner sectional areas and as a result get less dead load structures. Kwik Bond Polymers, which is one of the biggest polymer concrete overlays supplier in the United States, stated that the most important characteristics of a bridge deck overlay are high durability, impermeability to chlorides, and ability to bond to the substrate. According to Kwik Bond Polymers, the first full-scale overlay of Polyester Polymer Concrete was demonstrated in 1983 in Northern California, and this bridge was rated as element level one in 2010 Caltrans inspection report; it is the highest rating available for a nearly 30 year proven system. Portland cement concrete is a mixture of aggregate, sand, cement, and water. In contrast, PPC is a combination of aggregate and polyester resin which gives the feature of rapid cure time to PPC. This topic will be detailed described in chapter 3.
3


Figure 2.1 shows casting a conventional concrete bridge deck. A polymer concrete overlay deck is shown in figure 2.2.
Figure 2.1. Conventional concrete bridge deck. Adapted from equipmentworld.com,
2014.
Figure 2.2. Polyester polymer concrete overlay. Adapted from Polymer Paving on a Project in Sacramento, CA, 2009 by GOMACO World 37(2).
4


2.2 Bridge Deck Types
Bridge deck types are defined by the materials from which they are made and the manner in which those materials are fit together. Constructing a bridge deck that is both structurally sound and durable requires careful planning and preparation. Bridge decks are subjected to a group of stresses such as the spanning local flexural bending over the girders in the transverse direction which caused by the vehicle wheel loads. Also, in the composite sections, decks are subjected to longitudinal stresses caused by flexure along the span. On top of that, bridge decks are performing as horizontal diaphragms that are able to absorb and transfer the supports lateral loads, such as the wind or seismic loads. Reinforced concrete deck slabs are often used for prestressed concrete girder and steel girder bridges. Concrete deck slabs can be constructed on the site which is called cast-inplace deck or be delivered to the site as precast concrete panels. Typically, bridge deck includes steel reinforcement in both longitudinal and transverse directions, and it can contain post-tensioning tendons in case of using a precast deck. Steel plate decks, orthotropic steel decks, and wooden decks are other types of bridge decks which are often lighter and less time consuming to install. These types of decks are advantageous and competitive options for use in deck replacement and bridge rehabilitation projects when time is a concern. The main two deck types that polymer concrete is used are concrete and steel plates decks (Chavel, 2012).
2.2.1 Polymer concrete overlay with concrete deck
The two main purposes of using polyester polymer concrete overlays on concrete bridge decks are to create a smooth profile for riding surface and protect deck steel reinforcement. In order to be eligible to do its job, the overlay needs to be durable to
5


traffic wear, petroleum products, and weather conditions. In addition, the overlay should be bonded to the substrate, impervious to chloride ions, and free of cracks to keep all salt solution concentration away from the interface. These intrusional chemicals accelerate steel corrosion if they find their way to the unbonded interface (Maggenti, 2001).
The polyester polymer concrete overlay can be applied on new bridge decks or existing decks. For new concrete decks, while the concrete surface still plastic, a very rough texture must be created with an approved texturing device producing an acceptable
bondable surface. This matter will be discussed deeply in chapter 3.
Figure 2.3. Applying polyester polymer concrete overlay on Newburgh-Beacon Bridge new concrete deck in Beacon, NY. Adapted from Newburgh-Beacon Bridge work nearly done by Rife, J., October 2015, recordonline.com.
For existing concrete decks, the concrete deck surface shall be uniformly
lacerated to an approximate depth of 'A inch with no slick or polish surfaces in line with
AASHTO LRFD Bridge Design Specifications. Any deteriorated concrete shall be
completely removed to sound and natural concrete, where approved patching material
6


may be used to overhaul the deck. The surfaces of that reconditioned regions should also be scarified about Vi inch depth before placing the overlay. If the existing bridge deck contains wearing surface on top of it, the wearing surface needs to be removed before placing the overlay (Chavel, 2012).
Figure 2.4. Repairing and later applying overlay for Carlson bridge existing concrete deck in Illinois Route 7 over the Des Plaines River in Lockport, Illinois. Adapted from
Albin Carlson Contractors, 2010.
2.2.2 Polymer concrete overlay with steel deck
From technology advances to accelerated bridge construction, the usage of steel decks becomes desired more and more, because of their cost-saving advantages especially on long span bridges, their long service life, their light weight, and many more benefits. Canam Bridges, the leader company in orthotropic steel decks industry, stated that "Lighter and more flexible than its conventional concrete counterpart, orthotropic steel deck reduces assembly time in the field and limits disruptions to traffic." Polyester polymer concrete provides a smooth roadway for riding and protects the steel deck surface from traffic action and weather conditions (Jia et al., 2016).
7


For last two decades, orthotropic steel deck has been widely used in long bridges. The orthotropic deck is composed of a steel stiffened plate and overlay. The bottom of the plate is stiffened with steel stiffeners that designed according to the panel dimensions and traffic loads.
The historic landmark, Brooklyn Bridge in New York crossing the East River is an example of that taype of bridges, where in 2013 its deck was replaced by orthotropic steel deck as figures 2.5 and 2.6 below show.
Figure 2.5. Preparing Brooklyn Bridge orthotropic steel deck in the shop. Adapted from
Canam Group Inc., 2016.
Figure 2.6. New Brooklyn Bridge orthotropic steel deck. Adapted from Canam Group
Inc., 2016.
8


2.2.2.1 Using zinc-rich primer on steel decks.
Corrosion is a serious problem could steel bridges be exposed to, where they may lose their function after some years and before their service life end. Any steel structure will oxidize if it left undefended or ineffectively protected from the natural environment. Corrosion needs many years to grow obvious deterioration which significant enough to brings the attention. Corrosion is addressed, and a protection plan is assigned through the design phase.
Corrosion has played a notable role in tragic collapses, where Silver Bridge in Point Pleasant, WV was collapsed in 1967 at 39 years old causing 46 deaths and the Mianus River Bridge in Greenwich, CT was collapsed in 1983 at only 25 years old. Both bridges have been collapsed due to corrosion (Kogler, 2012).
Nowadays, the high-performance coating, zinc-rich primer, is used widely as the main corrosion protection component in steel bridges, where the majority of state highway departments required using zinc-rich primer based coating system (Albrecht, 2003). Transportation Research Board did a survey in 1996 and found that 42 of 54 bridge agencies specify zinc-rich primers for new construction (Appleman, 1997).
For steel deck with overlay, all surfaces that will have interaction with the polymer concrete should be cleaned, generally, by using abrasive blasting to provide a spotless surface for coating. Then, the plate deck is painted with zinc primer and left to dry before applying the polyester polymer concrete layer. Just like the concrete deck, one coat of prime coat shall be uniformly applied to the prepared steel surfaces immediately before placing the polyester polymer concrete. This prime could be mixed with the zinc primer and applied as one coat.
9


2.3 Bond Strength
Chemists define the bond strength as the degree to which each atom is joined to another. It is the resistance to separation of two materials in contact with shared surface area. On the other hand, engineers describe it by means of the amount of adhesion between attached surfaces measured in terms of the stress required to separate the adherent material from the substrate to which it is bonded.
Better bond strength between overlay and substrate is better performance of composite deck, where in full bonded surface, the stress distribution between the two layers is different than what it is in no bond condition. As visible in figure 2.7, the fully bonded interface has no any stress discontinuity, while the unbonded interface has high stress concentration. In partially bonded case, which is the typical more realistic case, there is some of the stress concentration at the interface.
Asphalt Binding Partially-Bonded Aggregate Interlocking
(Low Temperature) (for Real Pavement) (High Temperature)
Figure 2.7. Bonding stress distribution. Adapted from Numerical and Experimental Analysis for the Interlayer Behavior of Double-Layered Asphalt Pavement Specimens, 2011, pp. 14. ASCE Journal of Materials in Civil Engineering, 23(1).
10


Sort of adhesion is developed due to using adhesive in the overlay mixture which is the key to a good bond. However, the bond is high influenced by the surface condition and the physical and chemical properties of the repair material and the substrate concrete. Surface preparation is a significant factor that can reduce the bond strength, where equipment, material, technique, and procedures should be used to prepare surfaces for overlays. Surfaces need to be prepared by removing all material that may act as a bond breaker between the surface and the polymer concrete such as dust, grease, and other loose particles. The moisture is another factor effects the bond directly; it is a significant construction and quality problem for polymer applying for the reason that any drops of water will have an opposing reaction to the bonding strength between the polymer concrete and the cement causing failure at the interface. Also, a rough concrete surface should be provided by protruding the aggregate. Typically, sandblasting or abrasive blasting is used to rough the concrete surface. Figure 2.8 shows the interface between two layers, concrete as a base and polymer as an overlay.
Figure 2.8. Diagram of overlay and substrate.
11


In practice, the most common test method to evaluate bond strength is the pull-off test, where direct tension force is applied perpendicular to a piece of the overlay that has an interface area with substrate till the failure occurs. The bond strength is defined as the force that the failure occurred at divided by the interface area. (Bonaldo et al., 2005).
Even though shear strength is a major factor for bonding, the pull-off test is the most used bond test method due to its ease to use in the field.
The pull-off test is done in steps in the field. After both overlay and substrate are cured and ready to be in service, the test is done in below steps:
Cores are drilled at 90-degree angle through the overlay and about an inch into the substrate to ensure that we get a good bond area. Usually, 2 inches diameter cores are used.
Steel or aluminum disks are prepared to be attached on top of the cores by cleaning them before use to prevent interfacial failures at the disk.
The surfaces of the samples are cleaned.
An adhesive, which should be stick on both steel or aluminum and polymer concrete and stronger than both overlay and substrate to avoid its failure first, is mixed.
The adhesive is applied to steel disk and surface of the sample to make sure that the adhesive wet on both surfaces.
The disks are pressed on the surface of the sample with a firm pressure without twisting if the steel disk twist that will increase the interface failure.
12


All adhesive access around the steel disk should be removed to make sure that the
bond between only the steel disk face and surface of the sample to get a right result.
After the adhesive has cured, a pull-off tester machine is attached to the disk to pull the sample till break it.
Figure 2.9 is a sketch summarizes a general picture of a pull-off test.
Force
E p oxy
Overl ay Substrate
Figure 2.9. Concrete Pull-Off Testing Schematic. Adapted from Carrasquillo Associates LTD., 2015
2.3.1 Pull-off test failure modes
The failure could occur at four different levels whichever is the weakest. Depending on where the failure happens, at the overlay, substrate, or at the bonded interface, the tester gives the failure tensile force which represents the tensile strength of that layer. The failure could also happen at the epoxy used to attach the steel disk to the core. If the break happens perfectly at the bonded interface, the failure load will specify the actual strength of the bond between the substrate and the overlay. If the break occurs in the substrate, it is a sign that the bond strength is higher than the substrate itself, which
D i sc
13


indicates that the overlay and the interface are stronger than the concrete deck (Bonaldo et al. 2005). The four failure modes are illustrative in figure 2.10 below.
Failure in Bond failure at Failure in Bond failure at
substrate concrete/overlay overlay or epoxy/overlay
interface repair material interface
Figure 2.10. Failure modes of the Pull-off test. Adapted from Papworths Construction
Testing Equipment, 2016
14


CHAPTER III
EXPERIMENTAL WORK
3.1 Methods
The experiment was conducted to evaluate the bond strength between concrete or steel panels bridge deck and polymer concrete overlay. All of the concrete was compacted by a compacting rod to prevent any voids which lead to reduce the capacity of concrete, thus six concrete plates compaction and curing to be available for testing. On each pad, one test was conducted to evaluate the shear and tensile resistance. Cylinders were used for concrete compression tests. Each of three pads, four core samples were gathered; four cores were drilled 62.5 mm (2.5 in.) into the concrete base because of the Pull-off tensile tests are the most standard practice in the field. The results of the test were compared. The other three pads were prepared for the shear test by cutting each pad to four smaller pieces to get smooth sides and easy to handle on the test machine. Also, 1.5-inch polymer layer has been applied on top of four steel panels; two plates for pull-off test and the other two for fatigue test where each test has two phases.
3.2 Samples Preparation
3.2.1 Concrete samples
The concrete samples were made at the University of Colorado Denver Civil Engineering Lab by casting concrete panels as concrete bridge deck. One concrete mixture was used to cast six concrete panels within six batches. Each batch was used for one pad and one to three cylinders. Six square concrete pads named Pad 1 to Pad 6 with the dimensions 18 inches by 18inches (457 mm x 457 mm) and 3 inches thick (76 mm). The size of the pads was mainly influenced by the length required to drill for the Pull-off
15


test. Wood molds were made in preparation for concrete casting, and a group of 4X8 concrete test cylinders was cleaned and prepared for casting day as well. This is shown in figure 3.1.
Figure 3.1. Molds are ready before the casting day.
A ready concrete mix was used for the concrete samples. It was chosen for bridge deck concrete because it satisfies ACI requirements and exceeds the requirements of ASTM C928 R-3 for bridge deck repair and concertation. The brand of the mix is DOT Mix extended with up to 0.5-inch gravel as figure 3.2 shows one 80-pound bag of the product. A slump test was done before casting the samples to assess the workability of the fresh concrete. Meanderingly, this test was made to examine the water to cement ratio. The result of the test was a slump of 4 inches as that is shown in figure 3.3 below.
16


The manufacturing company, Quikrete, recommends in the products mixing instruction slump range of 3 to 7 inches.
|v|
44*i
DOT MlX'EXrENDED
PRE-Extende with -i2 (-umn) Gravel

Y.l.^COMMERCIAL CRUDE**
Quikrete
Just Add Water!
HUUWC: CAS CADS SERIOUS WW
----
NET WT-80 LB (36.3 kg)
Figure 3.2. The ready mixed concrete that was used in samples casting.
Figure 3.3. Slump test.
17


Compacting was done to all six samples, which were cast in two layers, by a compacting rod for twenty-five hit on each one ninth of the slab. Six concrete slabs and 8- 8 x 4 concrete cylinders were made.
Figure 3.4. Compacting the first layer of a concrete panel.
Figure 3.5. One of the six concrete slabs that have been cast.
18


3.2.2 Concrete curing
For the concrete mix that was used in this research, the curing was not required for 28 days because the mix is fast set concrete and it gains the major part of its strength in the first week of its age. Plastic sheets were used to cover the samples during first 24 hours to serve as weather protection and next day wet cotton sheets were added on top of the samples. These sheets were kept wet for one week of the panels age.
Figure 3.6. First day of concrete curing.
Figure 3.7. The curing process at seven days old.
19


3.2.3 Steel plates preparation
Part of this research is to examine the bonding strength between polymer concrete and steel deck panels. Thus, four steel plate samples were prepared to represent a steel bridge deck. The plates were cleaned and grinded to remove all the rust and accumulated dirt. Two plates were cut to fit them in the machine that will be used for fatigue test. Figure 3.8 displays one of the four plates after all clean process have been done.
Figure 3.8. Steel plate a day before applying the polymer overlay.
3.2.4 Surface preparing
For concrete samples, in order to achieve good bonding between the concrete and the overlay, a rough concrete surface should be provided. Typically, sand blasting is used to rough the concrete surface, but because that equipment is not available, a hammer and chisel were used to prepare the surfaces of six samples to the overlay. That process was
20


necessary to continue until some of the surface aggregates can be seen to make the bonding better.
Figure 3.9. The panel's surface after it has been roughed up.
For steel plates, two phases of tests were planned to run on steel samples; one phase when zinc primer, the main corrosion protection component, was applied to the steel and one phase without zinc primer. The reason for testing two types of steel surfaces is to see how the bond is influenced by the zinc primer coat. Figure 3.10 demonstrates the process of applying the zinc primer coat on one of the two plates.
21


3.2.5 Applying polymer concrete layer
On a sunny day with a temperature of 65 F, all polymer samples were accomplished. A specialist had come to the university lab to help out with the process. The mixture was arranged in four batches because of the short set time period, where after the mix is ready, the tester has only 20 minutes to finish applying all the mixture.
Figure 3.10. Appling zinc primer coat on the steel plate surface.
The polyester polymer used for this work was PPC-1121 from Kwik Bond which specifically designed to seal bridge deck surfaces for both overlay and patching systems. The mixture was prepared as follow:
22


Two 50 lb. bags of B-l 1 sand and one 50 lb. bag of B39 rock aggregate were used for each batch. Two gallons of a polyester binder resin (12% of aggregate weight), which acts as the binder of the mixture contents, were mixed with 10 fl oz (2% by weight of the resin) of DDM 9 catalyst, which is the responsible for hardening. The mixture was added to a clean mortar mixer immediately. Then B-39 rock was added first to reduce mixer splashing before adding the two bags of B-l 1 sand. Whole contents were mixed for about 2 minutes before dumping everything into a large PVC tub.
Figure 3.11. Polyester mixture preparing.
Before applying the overlay, KBP 204 primer from Kwik Bond was distributed onto concrete panels surface by a paint brush. The primer was prepared by mixing 1 gallon of KBP 204 healer/sealer primer with three fluid ounces of 6% Cobalt Drier (Dark Blue Material) and stirred for 10 seconds. Then three fluid ounces of Cumyl Hydro Peroxide were added and stirred for 30 seconds. Immediately after that, the entire pail contents were applied onto the surface. This is very fluid and will wet the surface quickly. KBP job is to re-bonds cracks in Portland cement concrete base and promotes adhesion to the polyester polymer concrete.
23


Figure 3.12. Polymer applying process.
It is worth mentioning that this material does not need to curing as the regular concrete and it could be put in service after two hours.
At the end of that day all samples were accomplished, and they are as following:
6 panels of 18x18x1.5 inches polymer concrete on top of 3-inches concrete.
2 panels of 18x18x1.5 inches polymer concrete on top of 3/8-inch thick steel plates with zinc primer.
2 panels of 18x18x1.5 inches polymer concrete on top of 3/8-inch thick steel plates without zinc primer.
4 polymer concrete beam prisms (16x4x3 inches).
6 polymer concrete 8x4 cylinders in additional to 4 cylinders of 8x4 which were received from Colorado Department of Transportation 4 months ago.
8 regular concrete 8x4 cylinders.
3.3 The Machines That Were Used
Five different tests were done conducted in the University of Colorado Denver Civil Engineering Lab using three type of machines. 220 kips MTS machine was used for compression, bending, and shear tests. 20 kip 89101MTS machine was used for two fatigue tests and DYNA Z Pull-Off Tester was utilized for the pull-off test.
24


25


CHAPTER IV
TESTING
4.1 Pull-off Test
The pull-off test consisted of 18 specimens (10 concrete specimens and 8 steel specimens) divided into 5 different pads. For concrete specimens, 2-inch diameter cores were drilled with 90-degree angle through the overlay and into the concrete by approximately one inch to make sure that failure would occur in all three layers, overlay, substrate and at the bonded interface. For steel specimens, the specimens distributed equally on two steel plates, one with zinc primer and one without, (2-inch) diameter cores as well were drilled through the overlay all the way down until scratching the steel plate to make sure that the specimens have a chance to fail at the interface if it is the weaker region. The surface of the specimens was grinded and cleaned to remove all fine gravel and laitance; all surfaces were rinsed with water and dried before (2- inch) diameter and (1- inch) thick aluminum discs were glued with 3M DP 405 epoxy adhesive. 24 hours were needed for the epoxy to cure before the pull-off test was conducted, when alOmm draw bolt, which was used to level the tester to ensure that the force was applied perpendicular to the specimens, was fixed to the test disc then fitted into the coupling in the bottom of DYNA Z Pull-Off Tester as seen in figure 4.1. The tester reads the tension pressure per one square inch applied on 2-inch diameter area along with maximum read. The read was increasing while the tester increases applied force manually. At the end of each test, the necessary pressure to detach the specimens from the pads and the test time were recorded. The 18 tests took place when the concrete was 30 days old, and the polymer concrete was 21 days old.
26


b ^
Figure 4.1. Attaching the test disc to DYNAZ Pull-Off tester.
Figure 4.2. A test disc glued to a concrete specimen.
Equations 4.1 and 4.2 were used to calculate the tension forces that were required to pull off each sample and load rate respectively. Table 4.1 lists the results for all 18 samples.
F = crXi4
(4.1)
Where:
F = Applied Tensile Force (lb)
= Tensile Stress (psi)
= Surface Area of Specimen (in2)
F
LR =
t
(4.2)
Where:
LR = Load Rate (lb/sec) t = test time (sec)
27


Steel with Zinc Steel Without Zinc Concrete Panels
ffl w w w o O O O O O td td td td > > > > Sample
to L to to -L to to
UJ UJ UJ UJ Core Cross
Ii 11 li 11 ii li 11 Ii 11 ii 11 ii 11 11 ii 11 11 11
4^ 4^ 4^ 4^ 4^ -t^ 4^ -t^ -t^ -t^ -t^ 4^ 4^ 4^ Section Area
Ii Ii Ii 11 ii li 11 Ii 11 ii 11 ii 11 11 ii 11 11 11
On On On On On On On On On On On On On On On On On On (in2)
On On On ^1 ^1 ^1 On On 4^ to -t^ UJ -t^ to UJ to Maximum
^1 NO NO to NO 11 11 ^1 On 11 to 00 O -t^ LAi ^1 Pressure read
H-1 O 00 4^ NO 00 to ^1 11 11 00 -t^ 00 o NO 11
(psi)
to to to to to to to 00 NO ^1 ^1 to to 00 U\
li 11 11 to 4^ to NO 11 UJ UJ to to o 11
o 00 On ^1 NO to ^1 NO 4^ 00 to o <1 * to to 00 On NO to ^1 Maximum Force
o Cl bo U) L bo bo o 11 bo o 00 On NO to to \j\ G\ 00 bo LJ <1 (lbs)
O ^1 UJ ii o o ^1 11 -t^ o ^1 oj
4^ 4^ ^1 O o -t^ On ^1 4^
On ^1 Average Tensile
NO On o strength
(psi)
CO CO C/3 C/3 C/3 C/3 C/3 p C/3 CO
jr+i Q jr+i Q jr+i Q (Tt) Q jr+i Q jr+i Q H+, C H+, C H+, C H+, C H+, C ft> c H+, C Q- H+, C H+, C
P < CD P < tn cd po < tn cd po < til cd po < CD po < tn cd po < tn cd po < tn cd P CT P-I CO P CT P-I CO P CT C/l P CT C/l P cr P-I t/3 p cr P-I c p cr P-I c g. tr p cr P-I CO p cr P-I to Failure Mode
S= 2- S= Hi* S= Hi* S= Hi* S= Hi* s= Hi* S= Hi* S= Hi* s= ot s= ct c of c ot c ct c ot c ot C co c ot c ct
>-t CO CD eg >-* po CD eg >-* po CD eg >-t po CD ^ >-t po cd eg >-t po cd eg >-t po o> eg >-t po o> eg 3 P CD P-CD 3 P CD P-CD ^ P CD S-CD ^ P CD S* CD ^ P CD P-CD ^ P CD P-CD ^ P CD P-CD ^ H-J. CD < CD ^ P CD P-CD ^ P CD P-CD
to 90 98 88 95 Complete Time
<1 00 00 to 00 G\ NO to o On (sec)
£ £ £ £ to to £ 9 NO NO 9 NO 9 00 00 ^1 9 Load Rate (lb/sec)
bo -P NO o Cl 00 NO bo o o NO On NO O o On 00 Cl o NO Ci o o to o NO ^1 o NO NO ^1 o to NO Table 4.1: Results of 18 pull off tests have been done for a verity of samples.


Figure 4.3. The ten concrete pull off samples.
29


C-l C-2
C-3 C-4 C-2
E-2
E-4 E-3 E-4
Figure 4.4. The eight steel pull off samples.
30


4.2 Compression Test
Compression tests were performed to determine if the polyester polymer concrete mixture as cast meets the requirement of the specified strength (/c') in the specification. This test was done by breaking cylindrical samples of the material in the 220-kips compression machine, the compressive strength was calculated by dividing the failure load by the cross sectional area as shown in equation 4.3 below. 4X8 cylindrical specimens. The two end surfaces of each 4X8 cylinder were grinded to make sure that they are level and vertical to the side of the cylinder.
fc = 2 (4.3)
A = r2n (4.4)
Where:
fc = Compressive Strength (psi)
P= Failure Load (lb)
A = Specimen Cross Sectional Area (in2) r = Specimen radius (in)
A = (2)2tt = 12.566 in2
Ten cylinders were tested at different ages to see if curing the specimen affects their compressive strength. CDOT provided with four samples that were tested at age 158 days. Three other samples were tested at 7 days old and three more at 24 days old.
31


Table 4.2. Results of compression tests.
Specimen # Age Failure Load Compressive Strength Maximum Compression Strain Load Rate Average Compressive Strength
kips psi in/in in/min psi
P 1-1 60.908 4847 0.00606 0.02
P 2-1 7 days 60.683 4829 0.00604 0.02 5002
P 4-1 66.907 5330 0.00617 0.02
P 1-2 61.311 4879 0.00621 0.02
P2-2 24 days 61.672 4905 0.00588 0.02 4877 4977
P 3-1 60.913 4847 0.00633 0.02
CDOT-1 63.145 5025 0.00578 0.02
CDOT-2 158 days 62.663 4987 0.00554 0.02 5051
CDOT-3 61.347 4882 0.00506 0.02
CDOT-4 66.733 5311 0.00554 0.02
Figure 4.15 shows how the samples were level and 90 angle with sample side. Figure 4.16 shows using laser extensometer from MTS to measure the deformation of the cylinder while it was being loaded. The next four figures present the three stages of cylinder samples and the failure shape for some samples.
37


Figure 4.15. Preparing polymer samples for compression test and checking the level of their surfaces by passing them on a concrete saw to grind any excess materials.
Figure 4.16. The laser extensometer measures the deformation of the cylinder.
38


Figure 4.17. Sample P1-1, the 7 days old, under the load and its failure shape.
Figure 4.18. Sample P3-1, the 24 days old, under the load and its failure shape.
39


Figure 4.19. Sample CDOT 1, the 158 days old, under the load and its failure shape.
Figure 4.20. Sample CDOT 3, the 158 days old, under the load and its failure shape.
40


4.3 Bending Test
The goal of this test is to examine the bending strength of the polymer concrete and compute its modulus of rupture. Four polymer concrete beam prisms (16x4x3 inches) were prepared 10 days before the test day. The test was performed as a four-point flexure test.
Force Force
* a k Loading pins 1
Ji V i k
Specimen
(k X)
Supporting pins
Figure 4.21. Four points flexure test setup. Adapted from substech.com, 2012.
A custom fabricated steel frame was built for the four-point bending test where the loading force was applied by means of two loading pins with a distance between them equal to a half of the distance between the supporting pins. The beam prisms were 16 inches long, so 14 inches between supporting pins and 7 inches between loading pins have been created. The test was run as a displacement control test by using the 220-kips machine with a load rate of 0.05 in/min for first beam and 0.02 in/min for last three beams. The first test was relatively fast, so the rate decreased for other tests. The reason for using displacement control test is safety consideration where a sudden brittle failure was expected. The deflection at the midspan (maximum deflection) was recorded, where an LVDT and a load cell were connected through different data recorder system than
41


what the machine has, and then the two output data were compared and combined according to their timeline.
Equations 4.4 and 4.5 were used to calculate the modulus of rupture and the bending moment respectively at each force recorded. All data were assembled and plotted in a Modulus of Rupture-Deflection graph one for each sample. Table 4.2 demonstrations the results for all 4 samples.
3 Fa
a
h d2
(4.3)
M
(4.4)
Where:
er = Modules of rupture (ksi)
F = Total force applied to the specimen by two loading pins (kips) a = The distance between the support and the close loading pin (in) b = Specimen width (in) d = Specimen thickness (in)
M= Bending moment (in-kip)
For our specimen, the width was 3 inches, the depth was 4 inches, and the distance between the support and the close loading pin was 3.5 inches.
42


Figure 4.26. Failure mode of beam sample #1.
Figure -/.27.The failure mode of beam sample #2.
45


Figure 4.28. The failure mode of beam sample #3.
Figure 4.29. The failure mode of beam sample #4.
46


Table 4.3: Results of all four bending tests.
Specimen# Maximum Load Maximum Moment Modulus of Rupture Maximum Deflection Load Rate Running Time
kips in-kips psi in in/min seconds
B-l 10.859 19.003 2375 0.087 0.05 104.8
B-2 9.375 16.407 2051 0.090 0.02 270.4
B-3 8.508 14.889 1861 0.086 0.02 257.2
B-4 8.680 15.190 1899 0.098 0.02 294.3
Average 9.356 16.372 2047 0.090
4.4 Shear Test
This test was performed to check whether concrete, overlay, or the interface is the weakest under shear stress. In other words, which element among them has the least shear strength.
To have the ability to run that test with MTS machine, the 18 in. xl8 in. pads were saw cut into smaller blocks using a concrete saw as seen in figure 4.30. Six blocks with the two layers (concrete and polymer overlay) were made, and two cuts were made through the entire overlay and approximately Vi in into the concrete. The elected piece was separated from the block at its sides and kept having a recognized interface area with the substrate. The interface area was 3X4 for two samples and 3X6 for other four samples.
At the same time, a steel framework was designed, chopped, and welded so that the machine is able to push assigned piece until detaching it from the block. The test was
47


run as a displacement control test by using the 220-kips machine with a load rate of 0.005 in/min.
Figure 4.30. Preparing shear samples and the framework for the test.
All six tests were done when the concrete was 29 days old, and the polymer concrete was 20 days old. At the same day, 8 concrete cylinders were tested for compressive strength to determine fc of the concrete that were used to make the pads at the same age when they were tested for shear. The concrete samples were named in a way to indicate the number of the mixture that the cylinder was made from and the number of the cylinder from that mixture, where letter C is for concrete, the first number specifies the concrete mixture, and the second number is the cylinder number from that concrete mixture. The cylinders were not perfectly flat on the top and bottom, so two pad caps and retainer rings were used to distribute the load uniformly onto the cylinders.
Concrete compressive strength test was performed simultaneously with some other tests, where at the same day of concrete cylinders test, the shear tests of six samples were done when the concrete age was 29 days. On the next day, all ten concrete pull off tests were completed with 30 days old concrete.
48


A brief summary of the results for concrete cylinders is shown in Table 4.4 below, and two cylinder tests are shown in Figures 4.35 and 4.36 as examples of the test.
Table 4.4: Concrete cylinders results.
Specimen# Maximum Load Compressive strength Load Rate Age
kips psi in/min Days
C 1-1 104.6 8326 0.03 29
C 1-2 120.7 9609 0.03 29
C 3-1 115.5 9190 0.03 29
C 3-2 115.5 9192 0.03 29
C 4-1 112.5 8953 0.03 29
C 5-1 103.7 8252 0.03 29
C 6-1 100.1 7965 0.03 29
C6-2 101.8 8105 0.03 29
Average 109.3 8699
The small shear blocks were made from cutting pads number 3 and 4. As the table above shows, the average of the compressive strength for all samples that were made from the same mixture of those two pads was 9112 psi.
For all six specimens, the concrete has failed before the overlay and the interface as Table 4.5 below presents all shear results. The table also shows the compressive strength for both concrete substrate and polymer concrete overlay. For polymer concrete compressive strength, the average of youngest six samples was used.
49


Table 4.5: Direct shear test results.
Sample # Fc' Concrete Fc' Polymer Overlay Failure Force Interface Area Failure Mode Shear Stress
Psi psi lb. in2 psi
Sl-1 9112 4940 5656 12 Concrete 471
Sl-5 9112 4940 6612 13.5 Concrete 490
S2-5 9112 4940 11000 18 Concrete 611
S3-5 9112 4940 8414 18 Concrete 467
S4-5 9112 4940 8489 18 Concrete 472
S5-5 9112 4940 8662 18 Concrete 481
Figure 4.31. Sample and frame set-up in the 220 kips MTS machine.
50


Figure 4.32. Sample Sl-1 when the determined area got separated by direct shear.
Figure 4.33. Sample S 5-5 when the determined area got separated by direct shear.
51
wibit'


Figure 4.34. All shear samples after all direct shear tests have done.
Figure 4.35. Concrete cylinder sample C 3-1 during test it for compressive strength.
52


Figure 4.36. Concrete cylinder sample C 6-1 during test it for compressive strength.
Figure 4.37. Six concrete cylinder sample failure shape under compression test.
53


4.5 Fatigue Test
Fatigue is the weakening of a material that is subjected to repeated loading and unloading where the structural damage occurs due to that cyclic loading. The nominal load values that cause such damage may be much less than the strength of the material.
The goal of this test is to examine the bond strength between polymer concrete overlay and a steel plate, as a sample of a steel deck bridge, under cyclic loading. Two specimens (18in. X 18in.) were prepared 18 days before first test day as mentioned in chapter 3 where the overlay layer was 1.5 inches thick. The steel plate sample has zinc primer to see the response of the bridge deck when a corrosion protection is used versus using plain steel.
The two tests were run through 20 kips MTS machine where a special steel framework was designed to place the 2.25 square inches sample under the machine safely as shown in figure 4.38. The steel frame was made to handle the sample as a simply supported slab. The plate was a bit larger than the overlay layer to provide extra space for the supports, but the interface area was only 2.25 square inches. The goal was to run as many cycles as possible in order to achieve some failure point. Thus, each test was run through a weekend to get that much of cycles. Two cycles per second were applied, and the test was performed as a displacement control test where the machine was pushing the sample by applying 4" diameter load at the slab center for 0.1 inch downward deflection and release it back to its normal level and push it again for another cycle and so on. The MTS system was recording the load, deflection, and time twenty times for each cycle to draw a smooth curve at the end.
54


The sample with zinc is the first sample tested. The test lasted for 40 hours recording about 300,000 cycles. Cracks were started propagating from the center after 171,655 cycles. The second sample had same cracks, but the cracks appeared earlier than
the sample with zinc, where first visible crack was recorded at cycle number 89,033.
Figure 4.38. Attaching a framework to the 20-kip machine.
Figure 4.39. Both samples during their tests.
55


Table 4.6 below briefly summarizes the most prominent events that have occurred
during both tests.
Table 4.6: Fatigue tests events summary.
Sample Event 1st Sample With zinc primer 2nd Sample Without zinc primer
Sample relaxation When the test started, 19.5 kips was needed to create 0.1 inch vertical deflection at the slab center. After first 100 cycles, the load dropped from 19.5 to 17.5 kips (10 %) while maintaining the 0.1 inch displacement. When the test started, 18.3 kips was needed to create 0.1 inch vertical deflection at the slab center. After first 190 cycles, the load dropped from 18.3 to 17.2 kips (6 %) while maintaining the 0.1 inch displacement
Isolation the overlay from the plate At cycle 2500 the polymer concrete layer got separated from the steel plate at 4 inches from all 4 comers. At cycle 100,000 the isolation lines dilated to 6 inches. The lines stopped at 6 inches where they did not expand in the subsequent cycles. Only at one side of one comer, 3 inches separation line has developed after 28,223 cycles. At cycle 31,115, the overlay separated from the plate at 1.5 inches at the other side of the same corner.
First 15,000 cycles After first 15,000 cycles, the load dropped from to 13.4 kips, losing 23% of the load after the sample has relaxed. After first 15,000 cycles, the load dropped from to 12.6 kips, losing 27% of the load after the sample has relaxed.
First crack At cycle 171,655, the first crack has emerged from the load point heading to the edge at the midspan with 10.1 kips load. At cycle 89,033, the first crack has emerged from the load point heading to the edge at the midspan with 8.17 kips load.
56


Table 4.6, continued
Sample Event 1st Sample With zinc primer 2nd Sample Without zinc primer
Crack from edge to edge The cracks quickly moved to the edges, where after few cycles from crack appearance, it reached the edge. At cycle 171,661, one crack along the slab in transverse direction has reached both unsupported edges. The cracks quickly moved to the edges, but slower than what happened in the first sample. At cycle 89,600, one crack along the slab in transverse direction has reached both unsupported edges.
Final configuration After transverse crack, the specimen got weaker and weaker. More cracks started to appear, and the load kept dropping. Cracks at supported edges have recorded at cycle 173,601. After transverse crack, the specimen got weaker and weaker. More cracks started to appear, and the load kept dropping. Cracks at supported edges have recorded at cycle 94,437.
After about 250,000 cycles, the load seemed to keep its value at 8.5 kips, where in the next 60,000 cycles it dropped by only 500 lbs. After about 140,000 cycles, the load seemed to keep its value at 6.9 kips, where in the next 80,000 cycles it dropped by only 400 lbs.
Stopping the test The test has been stopped after 310,000 cycles of 43 hours of running. The final load was 8 kips, which is 46% of starting load. The test has been stopped after 220,000 cycles of 30.5 hours of running. The final load was 6.5 kips, which is 38% of starting load.
57


fill*
V- 2nd Sample ^ without zint
1st Sample| with zinc
Figure 4.42. Both samples after they have been tested.
59


CHAPTER V
DISCUSSION
5.1 Pull-off Test Results
The compressive strengths for both overlay and concrete substrate have been examined Concurrently with the pull-off test.
For concrete substrate, the compressive test took place a day before the pull-off test, when the concrete was 29 days old. Theoretically, at this age, the concrete supposed to reached its maximum compressive strength. Thus, at the pull-off test day, the concrete compressive strength was assumed to have the same strength at the day before.
For polymer concrete overlay, the compressive strength test was run when the polymer was 7, 24, and 158 days old, while the pull-off test was performed at the age of 21 days. The polymer concrete, unlike the regular concrete, does not need a long time for curing, where it gains most of its strength in first two hours.
As polymer concrete cylinder compressive strength test showed, there is no significant improvement in polymer concrete compressive strength from 7 to 158 days old. Thus, the average of the first youngest six cylinder compressive strengths was considered for the pull-off test; the six cylinder samples were made of the same polymer concrete mixture as the pull-off samples.
From Table 4.1, the results show that the concrete substrate fail in tension before either polymer concrete overlay or the interface, which means the polyester polymer concrete is stronger than the high strength concrete substrate and it is bonded to that substrate with strength larger than the required. This applies to 9 out of 10 samples. For
60


the tenth sample, Sample number A-3, the adhesive between the overlay and the test disc failed before any other sample material. This failure mode is not considered because it does not indicate the real strength of the sample.
In the second part of table4.1, the steel plates part, the results show that all eight samples had the same failure mode. These results indicate the tensile strength of the polyester polymer concrete, where the overlay failed before the interface in all 8 samples.
There is no one sample failed at the interface either in the plate with zinc primer or in the plate without zinc. As a result, we were not able to find the bonding strength between the polyester polymer concrete overlay and steel plate. However, the bonding strength is larger than polymer concrete overlay tensile strength even with applying zinc primer on the steel plate at the interface.
Table 5.1 is the result of comparing the tensile strengths for concrete and polymer concrete samples with their compressive strength. The table shows that in general, polymer concrete has relatively high tensile strength comparing to concrete tensile strength. This means cracks will develop in the concrete slab earlier than the polymer concrete overlay on a bridge deck.
Theoretically, the tensile strength of concrete is between 10 to 15% of the compressive strength. However, this ratio is affected by many factors such as aggregate size, curing time, sample age, and the test method. These factors might influence the tensile strength of the concrete since it contains high strength aggregate and some additives to improve the compressive strength. Also, the pull-off test method is not the typical method to examine concrete tensile strength where splitting tension test is the typical tensile strength test.
61


Table 5.1: Comparing the tensile and compressive strengths for concrete and polymer.
Sample Compressi ve strength (fc) (psi) Tensile Strength (Ft) (psi) Failure Mode Ft/fc *100% Average Ft/fc *100%
Concrete Panels A-l 8699 271 substrate failure 3% 3%
A-2 8699 359 substrate failure 4%
A-3 Neglected adhesive failure -
A-4 8699 348 substrate failure 4%
B-l 8699 404 substrate failure 5%
B-2 8699 388 substrate failure 4%
B-3 8699 421 substrate failure 5%
B-4 8699 311 substrate failure 4%
D-l 8699 267 substrate failure 3%
D-2 8699 432 substrate failure 5%
Steel Without Zinc C-l 4912 678 overlay failure 14% 14%
C-2 4912 613 overlay failure 12%
C-3 4912 719 overlay failure 15%
C-4 4912 794 overlay failure 16%
Steel with Zinc E-l 4912 723 overlay failure 15%
E-2 4912 698 overlay failure 14%
E-3 4912 690 overlay failure 14%
E-4 4912 671 overlay failure 145
* The average compressive strength of eight concrete cylinders tested a day before the pull-off test at the age of 29 days.
62


5.2 Compression Test Result
As mentioned earlier, the polymer concrete compression test was performed at different ages. Table 4.2 shows that the polymer concrete compressive strength did not improve much during last 151 days of its age, where the strength that has been gained during that period is less than 1% of the compressive strength at 7 days old. This indicated that the curing has no significant effect on polymer concrete compressive strength.
There was a little deviation in the compressive strength at 24 days, but it is a small amount, 125 psi less; which may be because of the odd read of sample P 4-1, which is about 500 psi more than its peers.
Figures 4.5 through 4.14 and Table 4.2 show that polymer concrete typically has significant ultimate strain compressing to regular concrete ultimate strain. The average ultimate strain was 0.00586 in/in. On the other hand, ACI-Code reports that the strain corresponding to ultimate stress is about 0.003 for normal concrete.
The test was displacement control test. Thus, it was possible to draw complete stress-strain curve as it is shown in the figures 4.5 through 4.14. The curves show that polymer concrete behaves more ductile than normal concrete.
It is interesting to note that the polymer concrete cylinders, unlike the concrete cylinders, remained cohesive after the failure, as Figures 4.17 through 4.20 show.
5.3 Bending Test Results
As the polymer concrete compressive tests show, the age of the samples does not affect the ultimate strength. Thus, the 10 days old beam samples should be performing as a full strength material.
63


The four samples showed small deflection during the test with sudden failure. The four samples behaved as a brittle material.
The maximum deflection was less than 0.1 inch for 14 inches span. From Table 4.3, the average modulus of rupture was 2047 psi. If the material were homogenous, that flexural strength would be the same as the tensile strength of the polymer. According to the pull-off test, the average tensile strength of the eight samples, that failed at the polymer, was 2194 psi, that is close to being homogeneous material, where the difference between the two results is only 6.7 %.
5.4 Shear Test Results
All six samples failed at the concrete layer; the concrete is the weakest component of the sample. Again, the polymer concrete shows good bonding strength with the concrete substrate.
The shear strength results were little uneven except S2-5 sample, which got extra 28% of the average strength. There is no clear reason for that. It was made from the same panel of three other samples, and it had the same interface area of three samples of the test. The average concrete shear strength of the five samples was 476 psi.
The most important result from this test for this paper is that the polymer concrete shear strength and the shear strength at the interface are greater than the concrete shear strength. This guarantees no failure at the polymer concrete overlay and the bonded interface of the deck before the deck itself fails.
5.5 Fatigue Test Results
Table 4.6 briefly summarizes most important events of both fatigue tests. A conclusion can be made from that table as follows
64


The polymer concrete showed a good performance to cyclic loading, where it kept taking load even after it cracked; first sample up to 46% of starting load and second sample up to 38% of its starting load.
More load was needed to deflect first sample as the same deflection as the second sample.
In general, the bonding strength for both samples was extremely good, where after all that many cycles, it was not possible to separate the polymer overlay from the steel plate at the interface even for the plate with zinc primer which showed some separation at the comers.
The zinc primer, corrosion protector, might affect the bonding strength, but the strength was more than enough to hold the polymer layer to the plate for 140,000 load cycles after the first crack.
The sample without zinc primer got the first crack earlier than the sample with zinc; it was two times faster. That might be because the thin zinc layer gave to the slab higher ductility than the slab without zinc, so the slab developed cracks later than the other.
The crack lines were formulated in a way that identical to the yield line theory of slabs, where the main crack was along the slab connected the two unsupported edges.
5.6 Final Conclusion
Polymer concrete has become more popular in bridge and precast industries with more and more research going on. Polymer concrete may not be the most cost efficient
65


option, but it has the advantage when the performance and construction time are essential.
Through the experimental data of this thesis, the main points are as follows:
Polymer concrete tensile strength is about 14 percent of its compressive strength (698 psi).
Polymer concrete tensile strength is greater than the highest tensile strength of the concrete that has been used in bridge decks and the bonding strength even higher.
The bonding strength is larger than 794 psi, which is higher than recommended 250 psi minimum to qualify for a polymer concrete overlay.
Maximum compressive strength is 5330 psi with an average of 4977psi.
Modulus of rupture is higher than the tensile strength.
5.7 Further Research
The shear strength between polymer concrete overlay and steel plate have not examined in this work. Thus, to have a complete idea about concrete polymer shear strength, such test is recommended to be done.
The main issue with polymer concrete is its cost due to the high cost of polyester resin. It is recommended to search for alternative material to replace the expensive resin such as recycle solutions.
66


REFERENCES
AASHTO, 2014. AASHTO LRFD Bridge Design Specifications, 7th Edition, American Association of State Highway and Transportation Officials, Washington, DC.
Albrecht, P. and Hall, T.T., 2003. Atmospheric Corrosion Resistance of Structural Steels. Journal of Materials in Civil Engineering, 75(1).
Appleman, B., 1997. Lead-based Paint Removal for Steel Highway Bridges. National Cooperative Highway Research Program Synthesis 251.
Bonaldo, E., Barros, J.A.O and Lourenco, P.B. (2005). "Bond Characterization between Concrete Substrate and Repairing SFRC Using Pull-Off Testing, International Journal of Adhesion and Adhesives, Elsevier, United Kingdom, Dec 2005, 463-474.
Chavel, B., 2012. Steel Bridge Design Handbook: Bridge Deck Design. The Federal Highway Administration, 17, pp. 3-14.
Jia, X., Huang, B., Chen, S. & Shi, D., 2016. Comparative investigation into field
performance of steel bridge deck asphalt overlay systems. KSCE Journal of Civil Engineering, 20(7), pp. 2755-2764.
Kim, H., Arraigada, M., Raab, C. & Parti, M., 2011. Numerical and Experimental Analysis for the Interlayer Behavior of Double-Layered Asphalt Pavement Specimens. ASCE Journal of Materials in Civil Engineering, 23(1), pp. 12-20.
Kogler, R., 2012. Steel Bridge Design Handbook: Corrosion Protection of Steel Bridges. Federal Highway Administration, 19.
Maggenti, R., 2001. Polyester Concrete in Bridge Deck Overlays Report. SFOBB East Spans Safety Project Skyway Structure.
67


APPENDIX A
PRODUCT DATA SHEET: PPCTM 1121
923 Teal Drive Benicia, California 94510
(866) 434-1772 (707) 746-7981 Fax contact@kwikbondpolymers.com
BRIDGE DECK & ROADWAY REHABILITATION SYSTEMS
PRODUCT DATA SHEET: PPC 1121
PRODUCT DESCRIPTION
PPC -1121 is Kwik Bond's polyester-based overlay and patching system specifically designed to seal bridge deck surfaces, develop high mechanical strengths, have the ability to pave variable depth cross sections, and return traffic quickly. PPC-1121 achieves over 4000 psi in compressive strength within 24 hours as well as over 800 psi in tensile strength. Because of its strength gain curve, traffic can be safely returned within 2 hours at temperatures of down to 40 E In direct adhesion testing to Portland cement concrete used for bridge deck applications, the failure mode is cohesion within the Portland cement concrete.
PPC -1121 has the following performance advantages:
PPC-1121 conforms to all latest draft specifications for polyester polymer concrete
PPC -1121 has high strength properties in both compression and tensile
PPC Binder Resin has a long history of performance (In use since 1983)
PPC -1121, when mixed and applied properly, can return traffic safely within 1.5-3 hours at temperatures down to 40 F.
For todays congested bridges and highways, PPC 1121 is the right material for overlaying, patching, repairing, and rehabilitating Portland cement concrete, latex modified concrete, or silica fume modified concrete.
:wikBond
POLYMERS
SPECIAL FEATURES
Low viscosity for easy mixing
KBP 204 "healer/sealer" primer re-bonds cracks in Portland cement concrete base and
promotes adhesion to the polyester polymer concrete
Superior adhesion to Portland cement concrete, latex modified concrete, silica fume
concrete even under damp conditions
Rapid cure and strength development, traffic can be returned within 2 hours of finishing
Excellent finishing and sealing characteristics
Superior abrasion resistance to chains and studded snow tires
kwikbondpolymers.com
68


PRODUCT DATA SHEET: PPC1121
KWIK BOND POLYMERS, LLC
PHYSICAL PROPERTIES PPC Binder Resin
Specific Gravity 1.05-1.10
Weight per gallon (resin binder only) 9.0-9.4
Viscosity (ASTM D2196) 75-200 cps
Flash Point (Seta flash) 90 F
Tensile Strength (ASTM D-638, *A") >2500 psi
Tensile Elongation (ASTM D-638, V4") 35%, min.
Silane Coupler 1.0% min by weight of polyester resin
Styrene content 40-50%
Meets CARB (California Air Resources Board) Rule 1162
PHYSICAL PROPERTIES PPC Composite
Cured Density (ASTM C-138) 134-136 Ibs/cf
Adhesion (Cal-Trans Test Method 551) >500 psi
Combined average moisture absorption of the aggregates is less approximately 1%. Crushed aggregate particle is less than 45%.
TYPICAL AGGREGATE GRADATION*
Screen Size % Pass inti
3/8" 100
No. 4 70
No. 8 50
No. 16 44
No. 30 30
No. 50 5-20
No. 100 1
No. 200 T
APPLICATION
Surface Preparation: Shot-blasting is the preferred method of preparation for bridge deck overlays. The goal of surface preparation is to remove all deleterious dirt, asphalt and curing compounds from the deck surface as well as expose aggregate. Sandblasting, scarifying, chipping, or other cleaning processes are required to provide proper surface preparation for a long-lasting polymer overlay and/or patching system. Unsound concrete areas should be located by using a chain-drag or hammer. The unsound areas must be removed until a sound concrete base is established.
Patching: Patch all unsound bridge deck concrete with PPC 1121. Patches can be filled to 3" depth and more. Deep areas up to 5' x 5 can be deep without normally impacting stiffness of the PCC bridge deck. Design engineers should consider the semi-rigid nature of Polyester Polymer Concrete in those calculations. If design factors require a rigid patch system, utilize high alumina concrete patch systems. Properly placed high alumina concrete patch systems may be overlayed with PPC 24 hours after placement.
KBP 204 Primer:
Mix 1 gallon KBP 204 "healer/sealer primer with 3 fluid ounces of 6% Cobalt Drier (Dark Blue Material). Stir for 10 seconds. Add 3 fluid ounces of Cumyl Hydro Peroxide and stir for another 30 seconds. Immediately empty the entire pail contents onto the PCC surface. Application rate ranges from 70-100 sf/gal depending on porosity and surface texture of the deck. Re-distribute the primer using a paint brush for small area or rollers, squeegees, brooms for larger areas, wet-out the entire surface of the area to be repaired. KBP 103/204
kwikbondpolymers.com
69


PRODUCT DATA SHEET: PPC 1121
KWIK BOND POLYMERS, LLC
is very fluid and will wet the surface quickly. The excess will rapidly build-up at the lowest points in the prepared area. Excess primer is undesirable. Apply primer carefully to have as little excess build-up as possible. Some build-up is unavoidable. Note: This mix design represents a starting point for anticipated temperatures of 70 F during daytime conditions. Modifications may be required for working under different temperature conditions or during night time application. For very warm temperatures, night time application should be considered. Reducing CHP levels to 111 oz per gallon during elevated temperatures should be evaluated. During cold night time application, both catalyst and accelerator concentrations will need to be increased.
PPC-1121 Mix:
To a clean 9 cubic foot mortar mixer, add 4 gallons of PPC Binder Resin. Add 7-12 fl oz of DDM 9[MEKP}. Note: for faster strength gain at low temperatures add .l-.4% Z Cure accelerator to resin. While mortar mixer is turning with PPC Binder Resin and catalyst, add 2-50 lb bags of B39 [S39 alternative} rock and 4-50 lb bags of B-ll sand. Rock can be added first to reduce mixer splashing. Mix for 2 minutes depending on temperature. Dump catalyzed patching compound into a wheelbarrow or similar transfer device. Immediately recharge mixer with proper volume of PPC Binder Resin. Continue mixing procedure ONLY if crew is ready for another mix. The agitating mortar mixer with Binder Resin only, without catalyst, will keep your mixer clean and reduce build-up. Mix design modifications are required for changes in temperature or nighttime application. Higher or lower catalyst additions may be required for meeting traffic control requirements. Temperature and application timing have a definite effect upon set time of the polyester polymer concrete and the ultimate return to service.
Volumetric mixers may also be utilized for high output applications. The utilization of volumetric equipment is almost essential for projects requiring rapid return to surface on major Interstate projects.
FINISHING
Mixed PPCM 1121 material is placed utilizing a vibratory strike off screed, a slip form paving machine, or standard hand finishing tools for smaller patches. For small areas an aluminum straight edge or a vibrastrike screed may be acceptable to develop appropriate surface finish and compaction. After strike-off to final surface grade, apply topping sand in slight excess plus mechanically texture the surface utilizing spring steel tines 1/8" deep spaced 1" apart. Typical work time is 30 minutes. UV light accelerates the set time. PPC- 1121 is best used at temperatures between 40-90F. Adjustments to catalyst and concentrations are necessary when working outside the optimum temperature range. Trial batches are required to determine work times and set time based on anticipated application temperatures, conditions, and lane closure timing.
STANDARD PACKAGING
PPC Binder Resin 4 gal pail, 55 gal drum, 4400 gal tank truck
B-ll Sand-50 lb bags, 2 ton Super Sacks
B-39 Rock 50 lb bags, 2 ton Super Sacks S-39 Rock 50 lb bags, 2 ton Super Sacks
kwikbondpolymers.com
70


PRODUCT DATA SHEET: PPC 1121
KWIK BOND POLYMERS, LLC
Top Sand 10 x 30- 50 lb bags
DDM-9-1 gallon bottles
KBP 103/204 primer- available in 4 gallon pails 50 gallon drums, 250 gallon Tote Tanks
6% Cobalt Drier- available in pre-packaged bottles, 1-gallon cans, 4 gallon pails
Cumyl Hydro Peroxide- available in 1-gallon bottles, or 4-gallon cases
Z Cure- pre-packaged bottles, 1 gal cans, 5 gal bottle
SAFETY
PPC 1121 system consists of polymer materials that have been used safely for over 20 years. However, there are certain safety issues that need to be readily understood. PPC Binder Resin is FLAM- MABLE! NO SMOKING is allowed! Fire extinguishers must be available as well as plans for emergency situations. Emergency situations are unlikely, but preparation is always SMART!
The KBP 103/204 primer is a three-component system. The 6% Cobalt Drier and the Cumyl Hydro Peroxide are INCOMPATIBLE materials. They must NEVER be mixed together by themselves! A FLASH FIRE WILL OCCUR! To safely mix the KBP 103/204 primer, follow the mixing instructions EXACTLY! Follow the mixing instructions outlined in this product data sheet and safety will be maintained.
For emergency situations, always have available clean water for accidental contact in the eyes, fire extinguishers, and emergency center addresses, phone numbers.
Wear protective clothing, eye protection, and chemical resistant gloves. Organic vapor respirators are not normally required. For individuals highly sensitive to chemical vapors, organic vapor respirators are suggested.
STORAGE
Aggregates, PPC Binder Resin, and KBP 103/204 should be stored in a cool, dry location and in their original containers. The shelf life for these materials stored at temperatures 80 F and below is 12 months. PPC Binder Resin and KBP 103/204 contain reactive polymers. At elevated temperature, storage shelf life is reduced. Store all bagged aggregates in a clean, dry location away from moisture. Aggregates must absolutely be protected from any moisture.
The technical data furnished is true and accurate to the best of our knowledge. However, no guarantee of accuracy is given or implied. We suggest that customers evaluate these recommendations and suggestions in conjunction with their specific application. Kwik Bond Polymers, LLC warrants its products to be free from manufacturing defects conforming to its most recent material specifications. In the event of defective materials, Kwik Bond Polymers, LLC's liability will be limited to the replacement of material or the material value only at the sole discretion of Kwik Bond Polymers, LLC. Kwik Bond Polymers, LLC assumes no responsibility for coverage, suitability of applicatioa performance or injuries resulting from use. 3-16-2014
kwikbondpolymers.com
71


Full Text

PAGE 1

MECHANICAL AND BONDING PROPERTIES OF POLYESTER POLYMER CONCRETE BRIDGE DECK OVERLAY b y MARWA FOAD MANHER AL ETHAFA B.S., University of Al Qadisiya, Iraq, 2010 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Civil Engineering Program 2016

PAGE 2

ii This thesis for the Master of Science degree by Marwa Foad Manher Alethafa has been approved for the Civil Engineering Program by Frederick Rutz Chair Chengyu Li Nien Yin Chang Date : December 17 2016

PAGE 3

iii Alethafa, Marwa Foad Manher (M.S., Civil Engineering) Mechanical and Bonding Properties of Polyester Polymer Concrete Bridge Deck Overlay Thesis directed by Associate Professor Chengyu Li ABSTRACT Polymer Concrete (PC) is a composite material made from a combination of mineral aggregate such as sand and gravel with a polymerizing monomer. First, it has been used in many different areas such as patching, overlays for highway pavements and bridge decks, and flooring. In recent years, polyester polymer concrete has become widely used i n bridge deck overlay This article will investigate the material properties of the composite section with a variety of tests including cylinder compress ion, be ams bond strength, shear tests, and fatigue tests. Additio nally, the results obtained from those tests are critical to know the mechanisms of polyester polymer concrete. The form and content of this abstract are approved. I recommend its publication Approved: Chengyu Li

PAGE 4

iv ACKNOWLEDGEMENT S There are numerous people that in one way or another have helped me make this research successful. First of all, I would like to thank Dr. Chengyu Li who has served as an advisor for t his research His enthusiasm and dedication have bee n really motivating and he has been very supportive in every stage of the research, from planning and prepar ation to testing and writing this thesis. I would like to thank Tom Thuis, Peter Sillstrop and Jac Corless at the University of Colorado Denver Electronic Calibration and Repair Lab for being so generous with me and training me on MTS machines. I would also like to express my gratitude to Sheila Cherry who is a regional manager at Kwik Bond Polymer s. Sheila has done a lot supporting and pushing this thesis forward; her presence to apply the overlay, providing all mixture materials, and offering all the equipment for the pull off test. At last my husband and fellow graduate student Hussein Jaaz who spent several days helping me with casting concrete, coring pull off samples and saw cutting the shear test pads

PAGE 5

v TABLE OF CONTENTS CHAPTER INTRODUCTION ................................ ................................ ................................ ...... 1 1.1 Overview ................................ ................................ ................................ .............. 1 1.2 Research Objectives ................................ ................................ ............................. 2 LITERATURE REVIEW ................................ ................................ ........................... 3 2.1 Introduction ................................ ................................ ................................ .......... 3 2.2 Bridge Deck Types ................................ ................................ ............................... 5 2.2.1 Polymer concrete overlay with concrete deck ................................ .............. 5 2.2.2 Polymer concrete overlay with steel deck ................................ .................... 7 2.3 Bond Strength ................................ ................................ ................................ ..... 10 2.3.1 Pull off Test Failure modes ................................ ................................ ........ 13 EXPERIMENTAL WORK ................................ ................................ ....................... 15 3.1 Methods ................................ ................................ ................................ .............. 15 3.2 Samples Preparation ................................ ................................ ........................... 15 3.2.1 Concrete samples ................................ ................................ ........................ 15 3.2.2 Concrete curing ................................ ................................ ........................... 19 3.2.3 Steel plates preparation ................................ ................................ ............... 20 3.2.4 Surface preparing ................................ ................................ ........................ 20 3.2.5 Applying polymer concrete layer ................................ ................................ 22 3.3 The Machines That Were Used ................................ ................................ .......... 24 TESTING ................................ ................................ ................................ .................. 26 4.1 Pull off Test ................................ ................................ ................................ ........ 26

PAGE 6

vi 4.2 Compression Test ................................ ................................ ............................... 31 4.3 Bending Test ................................ ................................ ................................ ...... 41 4.4 Shear Test ................................ ................................ ................................ ........... 47 4.5 Fatigue Test ................................ ................................ ................................ ........ 54 DISCUSSION ................................ ................................ ................................ ........... 60 5.1 Pull off Test Results ................................ ................................ ........................... 60 5.2 Compression test Result ................................ ................................ ..................... 63 5.3 Bending Test Results ................................ ................................ .......................... 63 5.4 Shear Test Results ................................ ................................ .............................. 64 5.5 Fatigue Test Results ................................ ................................ ........................... 64 5.6 Final Conclusion ................................ ................................ ................................ 65 5.7 Farther Research ................................ ................................ ................................ 66 REFERENCES ................................ ................................ ................................ ................. 67 APPENDIX A. P roduct D ata S heet : PPCTM 1121 ... .68

PAGE 7

vii LIST OF TABLES TABLE 4.1: Results of 18 pull off tests have been done for a verity of samples. ..................... 28 4.2: Results of compression tests. ................................ ................................ ................. 37 4.3: Results of all four bending test s ................................ ................................ ............ 47 4 .4: Concrete cylinders' results. ................................ ................................ .................... 49 4.5: Direct shear test results. ................................ ................................ ......................... 50 4.6: Fatigue tests events summary. ................................ ................................ ............... 56 5.1: Comparing the tensile and compressive strengths for concrete and polymer. ....... 62

PAGE 8

viii LIST OF FIGURES FIGURE 2.1 Conventi onal concrete bridge deck ................................ ................................ ........ 4 2.2. Polye ster polymer concrete overlay ................................ ................................ ......... 4 2.3. Applying polyester polymer concrete overlay on Newburgh Beacon Brid ge new concrete deck in Beacon, NY ................................ ................................ ................... 6 2.4. Repairing and later applying overlay for Carlson bridge existing concrete deck in Illinois Route 7 over the Des Plaines R iver in Lockport, Illinois ............................ 7 2.5. Preparing Brooklyn B ridg e orthotropi c steel deck in the shop ................................ 8 2.6. New Brooklyn bridg e orthotr opic steel deck. ................................ .......................... 8 2.7. Bonding stress distribution. ................................ ................................ .................. 10 2.8. Diagram of overlay and substrate. ................................ ................................ ......... 11 2.9. Concret e Pull Off Testing Schematic ................................ ................................ .... 13 2.10. Failure modes of the p ull off test ................................ ................................ ........... 14 3.1 Molds are ready before the casting day. ................................ ................................ 16 3.2. The ready mixed concrete that was used in samples casting. ................................ 17 3.3. Slump test. ................................ ................................ ................................ ............. 17 3.4. Compacting the first layer of concrete panel. ................................ ........................ 18 3.5. One of the 6 concrete slabs that have been casted ................................ ................ 18 3.6. First day of concrete curing. ................................ ................................ .................. 19 3.7. Curing process at 7 days old. ................................ ................................ ................. 19 3.8. Steel plate a day before applying the pol ymer overlay. ................................ ......... 20 3.9. The panel's surface after it has been roughed up. ................................ .................. 21 3.10. Appling zinc primer coat on the steel plate surface. ................................ .............. 22

PAGE 9

ix 3.11. Polyester mixture preparing. ................................ ................................ .................. 23 3.12. Polymer applying process. ................................ ................................ ..................... 24 3.13. DYNA Z Pull Off Tester. ................................ ................................ ...................... 25 3.14. MTS 810 load frame. ................................ ................................ ............................. 25 4.2. Attaching the test disc to DYNA Z Pull Off tester. ................................ ............... 27 4.1. A test di sc glued to a concrete specimen. ................................ .............................. 27 4.3. The 10 concrete pull off samples. ................................ ................................ .......... 29 4.4. The 8 steel pull off samples. ................................ ................................ .................. 30 4.5. Stress strain curve for cylinder P 1 1 @ 7 days old. ................................ ............. 32 4.6. Stress strain curve for cylinder P 2 1 @ 7 days old. ................................ ............. 32 4.7. Stress strain curve for cylinder P 4 1 @ 7 days old. ................................ ............. 33 4.8 Stress strain curve for cylinder P 1 2 @ 24 days o ld. ................................ ........... 33 4.9. Stress strain curve for cylinder P 2 2 @ 24 days old. ................................ ........... 34 4.10. Stress strain curve for cylinder P 3 1 @ 24 da ys old. ................................ ........... 34 4.11. Stress strain curve for cylinder CDOT 1 @ 158 days old. ................................ .... 35 4.12. Stress strain curve for cylinder CDOT 2 @ 158 days old. ................................ .... 35 4.13 Stress strain curve for cylinder CDOT 3 @ 158 days old. ................................ .... 36 4.14. Stress strain curve for cylinder CDO T 4 @ 158 days old. ................................ .... 36 4.15. Preparing polymer samples for compression test and checking the level of their surfaces by passing them on a concrete saw to grind any excess materi als. ......... 38 4.16. The laser extensometer measures the deformation of the cylinder. ....................... 38 4.17. Sample P1 1, the 7 days old, under the load and its failure shape. ........................ 39 4.18. Sample P3 1, the 24 days old, under the load and its failure shape. ...................... 39 4.19. Sample CDOT 1, the 158 days old, under the load and its failure shape. ............. 40

PAGE 10

x 4.20. Sample CDOT 3, the 158 days old, under the load and its failure shape. ............. 40 4.21. Four points flexure test setup ................................ ................................ ................. 41 4.2 2. Modulus of Rupture Deflection graph of first beam sample. ................................ 43 4.23. Modulus of Rupture Deflection graph of second beam sample. ........................... 43 4.24. Modulus of Rupture Deflection graph of third beam sample. ............................... 44 4.25. Modulus of Rupture Deflection graph of fourth beam sample. ............................. 44 4.26. Failure mode of beam sample #1. ................................ ................................ .......... 45 4.27. Failure mode of beam sample #2. ................................ ................................ .......... 45 4.28. Failure mo de of beam sample #3 ................................ ................................ .......... 46 4.29. Failure mode of beam sample #4. ................................ ................................ .......... 46 4.30. Preparing shear samples and the framework for the t est. ................................ ...... 48 4.31. Sample and frame set up in the 220 kips MTS machine. ................................ ...... 50 4.32. Sample S1 1 when the determined area got separa ted by direct shear. ................. 51 4.33. Sample S 5 5 when the determined area got separated by direct shear. ................ 51 4.34. All shear sampl es after all direct shear tests have done. ................................ ........ 52 4.35. Concrete cylinder sample C 3 1 during test it for compressive strength. .............. 52 4.36. Concrete cylinder sample C 6 1 during test it for compressive strength. .............. 53 4.37. Six concrete cylinder sample failure shape under compression test. ..................... 53 4.38. Attaching a framework to the 20 kip machine. ................................ ..................... 55 4.39. Both samples during their tests. ................................ ................................ ............. 55 4.40. Graph for the first fatigue test shows the relation between applied load and the number of cycles. ................................ ................................ ................................ ... 58 4.41. Graph for the second fatigue test demonstrates th e relation between applied load and the number of cycles. ................................ ................................ ...................... 58

PAGE 38

27 Figure 4 2 A test disc glued to a concrete specimen. Figure 4 1 Attaching the test disc to DYNA Z Pull Off tester. Equation s 4.1 and 4.2 were used to calculate the tension forces that we re required to pull off each sampl e and load rate respectively. Table 4.1 lists the results for all 18 samples. # $ % & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & ( ) + Where: = Applied Tensile Force (lb) # = Tensile Stres s ( p si ) % = Surface Area of Specimen ,. + /0 1 & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & ( ) 2 Where: 34 = Load Rate ( lb /sec) 5&&"&& test time (sec)

PAGE 39

28 Table 4 1 : Results o f 18 pull off tests have been done for a verity of samples. S ample Core Cross Section A rea (in2) Maximum Pressure read (psi) M aximum F orce (lbs) Average Tensile strength (psi) F ailure M ode Complete Time of T est (sec) Load R ate ( l b/sec) Concrete Panels A 1 3.1416 271 851.374 356 substrate failure 95 8.962 A 2 3.1416 359 1127.834 substrate failure 142 7.942 A 3 3.1416 230 722.568 adhesive failure 88 8.211 A 4 3.1416 348 1093.277 substrate failure 130 8.410 B 1 3.1416 404 1269.206 substrate failur e 145 8.753 B 2 3.1416 388 1218.941 substrate failure 135 9.029 B 3 3.1416 421 1322.614 substrate failure 148 8.937 B 4 3.1416 311 977.038 substrate failure 98 9.970 D 1 3.1416 267 838.807 substrate failure 90 9.320 D 2 3.1416 432 1357.171 s ubstrate failure 156 8.700 Steel Without Zinc C 1 3.1416 678 2130.005 701 overlay failure 182 11.703 C 2 3.1416 613 1925.801 overlay failure 174 11.068 C 3 3.1416 719 2258.810 overlay failure 178 12.690 C 4 3.1416 794 2494.430 overlay failure 206 12.109 Steel with Zinc E 1 3.1416 723 2271.377 696 overlay failure 192 11.830 E 2 3.1416 698 2192.837 overlay failure 186 11.789 E 3 3.1416 690 2167.704 overlay failure 182 11.910 E 4 3.1416 671 2108.014 overlay failure 178 11.843

PAGE 40

29 Figure 4 3 The ten concrete pull off samples.

PAGE 41

30 Figure 4 4 The eight steel pull off samples

PAGE 42

31 4.2 Compression T est Compression test s were performed to determine i f the polyester polyme r concrete mixture as cast meets the requirement of the specified strength ( 6 7 8 ) in the specification. This test was done by breaking cylindrical samples of the material in the 220 kips compression machine. the compressive stren gth was calculated by dividing the failure l oad by the cross sectional area as shown in equatio n 4.3 b elow 4X8 cylindr ical specimens The two end surfaces of each 4X8 cylinder were grinded to make sure that th ey are level and vertical to the side of the c ylinder 6 7 8 9 % & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & ( ) : + % ; < & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & ( ) ( + Where: 6 7 8 & = Compressive S trength (psi) = = Failure Load (lb) % = Specimen Cross Sectional Area ,. + r = Specimen radius (in) % 2 < *2 ) >?? & & & & ,. Ten cylinders were tested at different ages to see if curing the specimen affects thei r compressive strength. CDOT provided with four samples that were tested at age 158 days. Three other samples were tested at 7 days old and three more at 24 days old.

PAGE 43

32 Figure 4 5 St ress strain curve for cylinder P 1 1 @ 7 days old. Figure 4 6 Stress strain curve for cylinder P 2 1 @ 7 days old. !"# $ %&%%'%' ()*(+ ,-. % /%%% 0%%% 1%%% (%%% 2%%% '%%% %&%%%% $ %&%%%1 $ %&%%/% $ %&%%0/ $ %&%%1( $ %&%%(* $ %&%%'2 $ %&%%*0 $ %&%%33 $ %&%//' $ %&%/1( %&1+3+ $ %&%/+/ $ %&%%// $ %&%%0+ %&'03+ $ %&%%/* $ %&%%%/ $ %&%%(+ Compressive Strength (psi) Strain (in/in) Stress Strain curve for cylinder P 1 1@ 7 days old !"# $ %&%%'%( ()*03 ,-. % /%%% 0%%% 1%%% (%%% 2%%% '%%% %&%%%% $ %&%%%0 $ %&%%%+ $ %&%%/* $ %&%%1% $ %&%%(0 $ %&%%2* $ %&%%++ $ %&%%3' $ %&%//+ $ %&%/1* $ %&%/'% $ %&%/*0 $ %&%0%2 $ %&%00+ $ %&%02% $ %&%0'3 $ %&%0*( $ %&%03/ Compressive Strength (psi) Strain (in/in) Stress Strain curve for cylinder P 2 1 @ 7 days old

PAGE 44

33 Figure 4 7 Stress strain curve for cylinder P 4 1 @ 7 days old. Figure 4 8 Stress strai n curve for cylinder P 1 2 @ 24 days old. !"# $ %&%%'/+ 2)11% ,-. % /%%% 0%%% 1%%% (%%% 2%%% '%%% %&%%%% $ %&%%// $ %&%%00 $ %&%%11 $ %&%%(2 $ %&%%2' $ %&%%'+ $ %&%%+* $ %&%%*3 $ %&%/%% $ %&%/// $ %&%/00 $ %&%/11 $ %&%/(2 $ %&%/2' $ %&%/'+ $ %&%/+* $ %&%/*3 $ %&%0%% $ %&%0// $ %&%001 $ %&%01( $ %&%0(2 $ %&%02' Compressive Strength (psi) Strain (in/in) Stress Strain curve for cylinder P 4 1 @ 7 days old !"# %&%%'0/ ()*+3 ,-. % /%%% 0%%% 1%%% (%%% 2%%% '%%% %&%%%% %&%%+( %&%%+0 %&%%'3 %&%%'+ %&%%'1 %&%%'% %&%%2' %&%%20 %&%%(* %&%%(( %&%%13 %&%%12 %&%%1/ %&%%0' %&%%0/ %&%%/2 %&%%/% %&%%%2 %&%%%% $ %&%%%2 $ %&%%/% $ %&%%/2 $ %&%%0% $ %&%%0( %&%%+2 Compressive Strength (psi) Strain (in/in) Stress Strain curve for cylinder P 1 2 @ 24 days old

PAGE 45

34 Figure 4 9 Stress str ain curve for cylinder P 2 2 @ 24 days old. Figure 4 10 Stress strain curve for cylinder P 3 1 @ 24 days old. !"# $ %&%%2** ()3%2 ,-. % /%%% 0%%% 1%%% (%%% 2%%% '%%% %&%%%2 %&%%%1 %&%%%% $ %&%%/% $ %&%%01 $ %&%%1* $ %&%%2( $ %&%%+/ $ %&%%** $ %&%/%( $ %&%/00 $ %&%/(/ $ %&%/'% $ %&%/+* $ %&%/3+ $ %&%0/+ $ %&%01' $ %&%022 $ %&%0+' $ %&%032 $ %&%1/2 $ %&%11( $ %&%120 $ %&%1'3 Compressive Strength (psi) Strain (in/in) Stress Strain curve for cylinder P 2 2 @ 24 days old !"# $ %&%%'11 ()*(+ ,-. % /%%% 0%%% 1%%% (%%% 2%%% '%%% %&%%%% $ %&%%%/ $ %&%%%2 $ %&%%/2 $ %&%%0* $ %&%%(0 $ %&%%23 $ %&%%+' $ %&%%30 $ %&%/%3 $ %&%/0' $ %&%/(' $ %&%/'2 $ %&%/*1 $ %&%0%/ $ %&%00/ $ %&%0(/ $ %&%0'% $ %&%0*% $ %&%1%% $ %&%1/3 $ %&%11* $ %&%12+ $ %&%1+( Compressive Strength (psi) Strain (in/in) Stress Strain curve for cylinder P 3 1 @ 24 days old

PAGE 46

35 Figure 4 11 Stress strain curve for cylinder CDOT 1 @ 158 days old. Figure 4 12 Stress strain curve for cylinder CDOT 2 @ 158 days old. !"# $ %&%%2+* 2)%02 ,-. % /%%% 0%%% 1%%% (%%% 2%%% '%%% %&%%%% $ %&%%%/ $ %&%%%( $ %&%%// $ %&%%0% $ %&%%1/ $ %&%%(( $ %&%%2* $ %&%%+( $ %&%%*3 $ %&%/%2 $ %&%/00 $ %&%/1* $ %&%/22 $ %&%/+1 $ %&%/30 $ %&%0// $ %&%01/ $ %&%021 $ %&%0+' $ %&%03* $ %&%100 $ %&%1(2 Compressive Strength (psi) Strain (in/in) Stress Strain curve for cylinder CDOT 1 @ 158 days old !"# $ %&%%22( ()3*+ ,-. % /%%% 0%%% 1%%% (%%% 2%%% '%%% %&%%%% $ %&%%%( $ %&%%%3 $ %&%%/( $ %&%%00 $ %&%%1/ $ %&%%(0 $ %&%%2( $ %&%%'+ $ %&%%*/ $ %&%%31 $ %&%/%' $ %&%//* $ %&%/03 $ %&%/1+ $ %&%/(0 $ %&%/(( $ %&%/(2 $ %&%/(2 Compressive Strength (psi) Strain (in/in) Stress Strain curve for cylinder CDOT 2 @ 158 days old

PAGE 47

36 Figure 4 13 Stress strain curve for cylinder CDOT 3 @ 158 days old. Figure 4 14 Stress strain curve for cylinder CDOT 4 @ 158 days old. !"# $ %&%%2%' ()**0 ,-.4 % /%%% 0%%% 1%%% (%%% 2%%% '%%% %&%%%% %&%%%% $ %&%%%0 $ %&%%%( $ %&%%/% $ %&%%/3 $ %&%%10 $ %&%%(' $ %&%%'/ $ %&%%++ $ %&%%31 $ %&%//% $ %&%/0+ $ %&%/(( $ %&%/'0 $ %&%/+* $ %&%/3' $ %&%0/1 $ %&%00+ $ %&%012 $ %&'+(1 Compressive Strength (psi) Strain (in/in) Stress Strain curve for cylinder CDOT 3 @ 158 days old !"# $ %&%%22( 2)1// % /%%% 0%%% 1%%% (%%% 2%%% '%%% %&%%%% %&%%%% $ %&%%%0 $ %&%%%* $ %&%%/2 $ %&%%02 $ %&%%1+ $ %&%%2% $ %&%%'( $ %&%%+3 $ %&%%31 $ %&%/%3 $ %&%/01 $ %&%/1* $ %&%/20 $ %&%/'+ $ %&%/*/ $ %&%/3( $ %&%0%+ $ %&%00/ $ %&%01+ $ %&%02/ $ %&%0'+ $ %&%0*0 $ %&%03* $ %&%1/( $ %&%10* $ %&%1(0 $ %&2'0( $ %&2'0( Compressive Strength (psi) Strain (in/in) Stress Strain curve for cylinder CDOT 4 @ 158 days old

PAGE 48

37 Table 4 2 Results of compression tests. Specimen # A ge Failure Load Compressive Strength Maximum Compression Strain Load Rate Average Compressive Strength kips psi in/in in/min psi P 1 1 7 days 60.908 4847 0.00606 0 .02 5002 4977 P 2 1 60.683 4829 0.00604 0.02 P 4 1 66.907 5330 0.00617 0.02 P 1 2 24 days 61.311 4879 0.00621 0.02 4877 P 2 2 61.672 4905 0.00588 0.02 P 3 1 60.913 4847 0.00633 0.02 CDOT 1 158 days 63.145 5025 0.00578 0.02 5051 CDOT 2 62.663 4987 0.00554 0.02 CDOT 3 61.347 4882 0.00506 0.02 CDOT 4 66.733 5311 0.00554 0.02 Figure 4.15 shows how the samples were level and 90 ¡ angle with sample side. Figure 4.16 shows using laser extensometer from MTS to measure the de formati on of the cylinder while it was being loaded The next four figures present the three stages of cylinder samples and the failure shape for some samples.

PAGE 49

38 Figure 4 15 Preparing polymer samples for compressi on test and checking the level of their surfaces by passing them on a concrete saw to grind any excess materials. Figure 4 16 The laser extensometer measures the deformation of the cylinder

PAGE 50

39 Figure 4 17 Sample P1 1 the 7 days old, under the load and its failure shape. Figure 4 18 Sa mple P3 1 the 24 days old, under the load and its failure shape.

PAGE 51

40 Fig ure 4 19 Sample CDOT 1, the 158 days old, unde r the load and its failure shape. Figure 4 20 Sample CDOT 3, the 158 days old, under the load and its failur e shape.

PAGE 52

41 4.3 Bending T est The goal of this test is to examine the bend ing strength of the polymer concrete and compute its modulus of rupture. Four polymer concrete beam prisms (16! 4 3 inches) were prepared 10 days before the test day. The test was performed as a four point flexure test. Figure 4 21 Four points flexure test setup. Adapted from substech.com, 2012. A custom fabricated steel frame was built for the four point bending test where the loading force was applied by means of two loading pins with a distan ce between them equal to a half of the distance between the supporting pins. The beam prisms were 16 inches long, so 14 inches between su pporting pins and 7 inches between loading pins have been created. The test was run as a displacement control test by u sing the 220 kips machine with a load rate of 0.05 in/min for first beam and 0.02 in/min for last three beams. The first test was relatively fa st, so the rate decreased for other tests The reason for using displacement control test is safety consideration where a sudden brittle failure was expect ed The deflection at the midsp an (maximum deflection) was recorded where a n LVDT and a load cell were connected through different data recorder system than

PAGE 53

42 what the machine has, and then the two output data were compared and combined according to the ir timeline Equation s 4.4 and 4.5 were used to calculate the modul u s of rupture and the bending moment respectively at each force recorded. All data were assembled and plotted in a M odulus of Rupture Deflection grap h one for each sample Table 4.2 demons trations the results for all 4 samples. # : & & @ A & & B & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & ( ) : + C 2 $ @ & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & & ( ) ( + Where: # = Modules of rupture (k si ) = T otal force applied to the specimen by two loading pins (kips ) @ The distance between the support and the close loading pin (in) b = S pecimen width (in) d = S pecimen thickness (in) M= Bending moment (in ki p) For our specimen, the width was 3 inches, the depth was 4 inches, and the distance between the support and the close loading pin was 3.5 inches.

PAGE 54

43 Figure 4 22 Modulus of Rupture Deflection graph of fi rst beam sample. Figure 4 23 Modulus of Rupture Deflection graph of second beam sample. !"# $ %&%*+1 .5 0&1+2 6-. %&%% %&2% /&%% /&2% 0&%% 0&2% %&%% $ %&%/ $ %&%0 $ %&%1 $ %&%( $ %&%2 $ %&%' $ %&%+ $ %&%* $ %&%3 $ %&/% Modules of rupture (ksi) Deflection (in) Modulus of Rupture Deflection graph of beam #1 !"# $ %&%3%/ .5 0&%2/ 6-. %&%% %&2% /&%% /&2% 0&%% 0&2% %&%% $ %&%/ $ %&%0 $ %&%1 $ %&%( $ %&%2 $ %&%' $ %&%+ $ %&%* $ %&%3 $ %&/% $ %&// $ %&/0 Modules of rupture (ksi) Deflection (in) Modulus of Rupture Deflection graph of beam #2

PAGE 55

44 Figure 4 24 Modulus of Rupture Deflection graph of third beam sa mple. Figure 4 25 Modulus of Rupture Deflection graph of fourth beam sample. !"# $ %&%*' .5 /&*'/ 6-. %&%% %&2% /&%% /&2% 0&%% 0&2% %&%% $ %&%/ $ %&%0 $ %&%1 $ %&%( $ %&%2 $ %&%' $ %&%+ $ %&%* $ %&%3 $ %&/% $ %&// $ %&/0 Modules of rupture (ksi) Deflection (in) Modulus of Rupture Deflection graph of beam #3 !"# $ %&%3* .5 /&*33 6-. %&%% %&2% /&%% /&2% 0&%% 0&2% %&%% $ %&%/ $ %&%0 $ %&%1 $ %&%( $ %&%2 $ %&%' $ %&%+ $ %&%* $ %&%3 $ %&/% $ %&// $ %&/0 Modules of rupture (ksi) Deflection (in) Modulus of Rupture Deflection graph of beam #4

PAGE 56

45 Figure 4 26 Failure mode of beam sample #1. Figure 4 27 The f ailure mode of beam sample #2.

PAGE 57

46 Figure 4 28 The f ailure mode of beam sample #3 Figure 4 29 The f ailure mode of beam sample #4.

PAGE 58

47 Table 4 3 : Results of all four bending tests Specimen# Maximum Load Maximum Moment Modulus of Rupture Maximum Deflection Load Rate Running T ime kips in kips p si in in/min seconds B 1 10.859 1 9.003 2 375 0.087 0.05 104.8 B 2 9.375 16.407 2 051 0.090 0.02 270.4 B 3 8.508 14.889 1 861 0.086 0.02 257.2 B 4 8.680 15.190 1 899 0.098 0.02 294.3 Average 9.356 16.372 2 047 0.090 4.4 Shear T est This test was p er formed to check whether concrete, overlay, o r the interface is the weakest under shear stress. In other words, which element among them has the least shear strength. To ha ve the ability to run that test with MTS machine the 18 in. x18 in pads were saw cut into smaller blocks using a concrete saw as seen in figure 4. 30 Six blocks with the two layers (concret e and polymer overlay) were made and two cuts were made through the en tire overlay and approximately in into the concrete. The elected piece was separate d from the block at its sides and kep t having a recognized interface area with the substrate. The interface area was 3X4 for two samples and 3X6 for other four samples. At the s ame time, a steel framework was designed, chopped, and welded so that the machine is able to push assigned piece un til detach ing it from the block. The test was

PAGE 59

48 run as a displacement control test by using the 220 kips machine with a load rate of 0.005 in/min. Figure 4 30 Preparing shear samples and the framework for the test. All six tests were done when the concrete was 29 days old and the polymer concrete was 20 days old. At the same day 8 concrete cylinders were tested for compressive strength to determi ne fc of the concrete that were used to make the pads at t he same age when t hey were tested for shear. The concrete samples were named in a way to indicate the number of the mixture that the cylinder was made from and the number of the cylinder from that mixture, where letter "C" is for concrete, the first number specifies the concrete mixture, and the second number is the cylinder number from that concrete mixture. The cylinders were not perfectly flat on the top and bottom so two pad caps and retainer r ings were used to distribute the load uniformly onto the cy linders Concrete compressive strength test was performed simultaneously with some other tests where at the same day of concrete cylinders' test, the shear t est s of six samples were done when the concrete age was 29 days. On the next day, all ten co ncre te pull off tests were completed with 30 days old concrete.

PAGE 60

49 A brief summary of the results for concrete cylinders is shown in T able 4.4 below and two cylinder tests are shown in F igures 4 .35 and 4.36 as examples of the test. Table 4 4 : Concrete cylinders' results. Specimen# Maximum Load Compressive strength Load Rate Age kips p si in/min Days C 1 1 104.6 8326 0.03 29 C 1 2 120.7 9609 0.03 29 C 3 1 115.5 9190 0.03 29 C 3 2 115.5 9192 0.03 29 C 4 1 112. 5 8953 0.03 29 C 5 1 103.7 8252 0.03 29 C 6 1 100.1 7965 0.03 29 C 6 2 101.8 8105 0.03 29 Average 109.3 8699 The s mall shear blocks were made from cutting pads number 3 and 4. As the table above shows, the average of the compressive strength for all samples that were made from the same mixture of those two pads was 9112 psi For all six s pecimen s, the concrete has failed before the overlay and the in terface as T able 4.5 below presents all shear results. The table also shows the compressive strength for both concrete substrate and polymer concrete overlay For polymer concrete compressive strength the average of youngest six samples was used

PAGE 61

50 Table 4 5 : Direct shear test results. Sample # Fc' Concrete Fc' Polymer Overlay Failure Force Interface Area Failure Mode Shear Stress P si psi lb. in2 psi S1 1 9112 4940 5656 12 Concrete 471 S1 5 9112 4940 6612 13.5 Concrete 4 90 S2 5 9112 4940 11000 18 Concrete 611 S3 5 9112 4940 8414 18 Concrete 467 S4 5 9112 4940 8489 18 Concrete 472 S5 5 9112 4940 8662 18 Concrete 481 Figure 4 31 Sample and frame set up in the 220 kip s MTS machine.

PAGE 62

51 Figure 4 32 Sample S1 1 when the determined area got separated by direct shear. Figure 4 33 Sample S 5 5 when the determined area got sep arated by direct shear.

PAGE 63

52 Figure 4 34 All shear s ample s after all direct shear tests have done. Figure 4 35 Concrete cylinder sample C 3 1 during test it for compressive strength.

PAGE 64

53 Figure 4 36 Concrete cylinder sample C 6 1 during test it for compressive strength. Figure 4 37 Six concrete cylinder sample f ailure shape under compression test.

PAGE 65

54 4.5 Fatigue T est Fatigue is the weakening of a material that is subjected to repeated loading and unloading where the structural damage occurs due to that cyclic loading. The nominal load values that cause such damage may be much less than the strength of the material. The goal of this test is to examine the bond strength between polymer concrete overlay and a steel plate, as a sample of a steel deck bridge, under cyclic load ing Two s pecimen s (18in. X 18in.) were prepared 18 days before first test day as mentioned in chapter 3 where the overlay layer was 1.5 inch es thick The steel plate sample has zinc primer to see the response of the bridge deck when a corrosion protection is used versus using plain steel. T he two tes ts were run through 20 kips MTS machine where a special steel framework was designed to place the 2.25 square inches sample u nder the machine safely as shown in figure 4.38. The steel frame was made to handle the samp le as a simply supported slab. T he plat e was a bit larger than the overlay layer to provide extra space for the supports but the interface are a was only 2.25 square inches. T he goal was to run as many cycles as possible in order to achieve some failure point. T hus, each test was run through a week end to get that much of cycles. T wo cycles per second were applied and the test was performed as a displacement control test where the machine was pushing the sample by applying 4" diameter load at the slab center for 0.1 inch downward deflection and release it back to its normal level and push it again for another cycle and so on. T he MTS system was recording the load, deflection, and time twenty times for each cycle to draw a smooth c u rve at the end.

PAGE 66

55 The sample with zinc is the first sample tested The test last ed for 40 hours recording about 300,000 cycles C racks were started propagating from the center after 171,655 cycles. The s eco nd sample had same cracks but the cracks appeared earlier than the sample with zinc wh ere first visible crack w as recorded at cycle number 89,033 Figure 4 38 Attaching a framework to the 20 kip machine. Figure 4 39 Both samples during their tests.

PAGE 67

56 Table 4.6 below briefly summari z es the most prominent events that have occurred during both tests. Table 4 6 : Fatigue tests events summary. Sample Event 1 st Sample With zinc primer 2 nd Sample Without zinc primer Sample relaxation When the test started, 19.5 kips was needed to create 0.1 inch vertical deflection at the slab center. After first 100 cycles, the load dropped from 19.5 to 17.5 kips (10 %) w hile maintaining the 0.1 inch displacement. When t he test started, 18.3 kips was needed to create 0.1 inch vertical deflection at the slab center. After first 190 cycles, the load dropped from 18.3 to 17.2 kips (6 %) w hile maintaining the 0.1 inch displacement I solation the overlay from the plate At c ycle 2500 the polymer concrete layer got separated from the steel plate at 4 inches from all 4 corners. At cycle 100,000 the isolation lines dilate d to 6 inches The lines stopped at 6 inches where they did not expand in the s ubsequent cycles. Only at one side of one corner, 3 inches separation line has developed after 28,223 cycles. At cycle 3 1,115, the overlay separated from the plate at 1.5 inch es at the other side of the same corner. First 15,000 cycles After first 15,000 cycles, the load dropp ed from to 13.4 kips losing 23% of the load after the sample has relaxed. After first 15,000 cycles, the load dropped from to 1 2.6 kips losing 27 % of the load after the sample has relaxed. First crack At cycle 171,655, the first crack has emerge d fr om the load point heading to t he edge at the midspan with 10.1 kips load. At cycle 89,033, the first crack has emerged from the load point heading to the edge at the midspan with 8. 1 7 kips load.

PAGE 68

57 Table 4.6, continued Sample Event 1 st Sample With zinc primer 2 nd Sample Without zinc primer Crack from edge to edge The cracks quickly moved to the ed ges, where after few cycles from crack appearance it reached the edge. At cycle 171,661, one crack along the slab in transvers e direction has reached both unsupported edges. The cracks quickly moved to the edges, but slower than what happened in the first sample. At cycle 89,600 one crack along the slab in transvers e direction has reached both unsupported edges. Final configuration After transvers e crack, the specimen got weaker and weaker. More cracks started to appear and the load ke p t drop ping. Cracks at supported edges ha ve recorded at cycle 173,601. After about 250,000 cycles, the load seemed to keep its value at 8.5 kips where in the next 60,000 cycles it dropped by only 500 lbs. After transvers e crack, the specimen got weaker and we aker. More cracks started to appear and the load ke p t dropping. Cracks at supported edges ha ve recorded at cycle 94,437. After about 14 0,000 cycles, the load seemed to keep its value at 6.9 kips where in the next 8 0 ,000 cycles it dropped by only 4 00 lbs Stopping the test The test has been stopped after 310,000 cycles of 43 hours of running. The final load was 8 kips which is 46% of starting load. The test has been stopped after 220,000 cycles of 30.5 hours of running. The final load was 6.5 kips wh ich is 38% of starting load.

PAGE 69

58 Figure 4 40 Graph for the first fatigue test shows the relation between applied load and t he number of cycles. Figure 4 41 Graph for the second fatigue test demonstrate s the rela tion between applied load and t he number of cycles % 0 ( /% /0 /( /' /* 0% 00 %4 /%4 0%4 1%4 (%4 2%4 '%4 +%4 *%4 3%4 /%%4 //%4 /0%4 /1%4 /(%4 /2%4 /'%4 /+%4 /*%4 /3%4 0%%4 0/%4 00%4 01%4 0(%4 02%4 0'%4 0+%4 0*%4 03%4 1%%4 1/%4 Compression Load (kips) x1000 Cycles % 0 ( /% /0 /( /' /* 0% 00 %4 /%4 0%4 1%4 (%4 2%4 '%4 +%4 *%4 3%4 /%%4 //%4 /0%4 /1%4 /(%4 /2%4 /'%4 /+%4 /*%4 /3%4 0%%4 0/%4 00%4 Compression Load (kips) x1000 Cycles

PAGE 70

59 Figure 4 42 Both samples after they have been tested

PAGE 78

67 REFERENCES AASHTO, 2014 AASHTO LRFD Bridge Design Specifications, 7th Edition, American Association of State Highway and Transportation Officials, Washing ton, DC Albrecht, P. and Hall, T.T., 2003. Atmospheric Corrosion Resistance of Structural Steels. Journal of Materials in Civil Engineering, 15 (1 ). Appleman, B., 1997. Lead based Paint Removal for Steel Highway Bridges. National Cooperative Highway Rese arch Progr am Synthesis 251. Bonaldo, E., Barros, J.A.O and Lourenco, P.B. (2005). "Bond Characterization between Concrete Substrate and Repairing SFRC Using Pull Off Testing," International Journal of Adhesion and Adhesives Elsevier, United K ingdom Dec 2005, 463 474 Chavel, B., 2012. Steel Bridge Design Handbook: Bridge Deck Design The Federal Highway Administration 17, pp. 3 14. Jia, X., Huang, B., Chen, S. & Shi D., 2016 Comparative investigation into field performance of steel bridge deck aspha lt overlay systems. KSCE Journal of Civil Engineering, 20 (7), pp. 2755 2764. Kim, H., Arraigada, M., Raab, C. & Partl M., 2011. Numerical and Experimental Analysis for the Interlayer Behavior of Double Layered Asphalt Pavement Specimens. ASCE Journal of Materials in Civil Engineering 23(1), pp. 12 20. Kogler, R., 2012. Steel Bridge Design Handbook: Corrosion Protection of Steel Bridges. Fede ral Highway Administration, 19. Maggenti, R., 2001. Polyester Concrete in Bridge Deck Overlays Report. SFOBB Eas t Spans Safety Project Skyway Structure

PAGE 79

68 APPENDIX A PRODUCT DATA S HEET: PPCTM 1121

PAGE 80

69

PAGE 81

70

PAGE 82

71