Citation
Mechanical and bonding properties of polyester polymer overlay on steel bridge deck under thermal conditions

Material Information

Title:
Mechanical and bonding properties of polyester polymer overlay on steel bridge deck under thermal conditions
Creator:
Alkhuraiji, Ahmed Suliman Abdullah
Place of Publication:
Denver, CO
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 engineering
Committee Chair:
Li, Chengyu
Committee Members:
Rutz, Frederick R.
Nogueira, Carnot

Notes

Abstract:
Polyester Polymer Concrete (PPC) is a composite material made of sand, aggregate, polyester resin, which serves as a binder. PPC is generally used on as an overlay on bridge decks and to repair bridge decks. This paper investigated the bond between steel decking material and primers used to bond PPC to steel decks. The bond strength of two types of primers was investigated: zinc primer and primer without zinc. Additionally, the tensile strength of PPC was inspected at various temperatures from 70ËšF to 164ËšF. A theoretical equation was developed to determine the tensile strength of PPC at any temperature between 70ËšF and 164ËšF, using the compressive strength of PPC at 70ËšF.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
Copyright Ahmad Suliman Abdullah Alkhuraiji . 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 is only available as the following downloads:


Full Text

PAGE 1

MECHANICAL AND BONDING PROPERTIES OF POLYESTER POLYMER OVERLAY ON STEEL BRIDGE DECK UNDER THERMAL CONDITIONS by AHMAD SULIMAN A BDULLAH ALKHURAIJI B.S., King Abdulaziz University, 201 4 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 201 8

PAGE 2

ii © 201 8 AHMAD SULIMAN A BDULLAH ALKHURAIJI ALL RIGHTS RESERVED

PAGE 3

iii This thesis for the Master of Science deg ree by Ahmad S uliman A bdullah Alkhuraiji has been approved for the Civil Engineering Program b y Chengyu Li, Chair Frederick Rutz Carnot Nogueira Date : May 12, 2018

PAGE 4

iv Alkhuraiji, Ahmad Suliman Abdullah (M.S., Civil Engineering) Mechanical and Bonding Pr operties of Polyester Polymer Overlay o n Steel Bridge Deck Under Thermal Conditions Thesis directed by Associate Professor Chengyu Li ABSTRACT Polyester Polymer Concrete (PPC) is a composite material made of sand, aggregate, polyester resin, which serves as a binder. PPC is generally used on as an overlay on bridge decks and to repair bridge decks. This paper investigated the bond between steel decking material and primers used to bond PPC to steel decks. The bond strength of two types of primers was inves tigated: zinc primer and primer without zinc. Additionally, the tensile equation was developed to determine the tensile strength of PPC at any temperature . T he form and content of this abstract are approved. I recommend its publication. Approved: Chengyu Li

PAGE 5

v ACKNOWLEDGEMENTS First and foremost, all thanks to Allah the most Gracious and most Merci ful for helping me to finish this thesis. Secondly, I would like to take this opportunity to express my sincere gratitude to Prof. Chengyu Li, P.E., S.E., Ph.D., who has been associated with this thesis and has helped me with his unflagging support, valuab le experience and continuous encouragement throughout this experimental study. I give great thanks to my country, Saudi Arabia, for the scholarship that has allowed me to pursue my higher education at the University of Colorado Denver. I would also express my gratitude to Tom Thuis and Jac Corless for their generous assistance with the sample preparations and tests. Also, I would like to express my thanks to Sheila Cherry, who provided materials, including Polyester Polymer Concrete, primers, and the pull o ff test machine. She taught me how to mix the materials according the standards of Kwik Bond Polymers, Inc., and helped with the material preparations. Finally, my thanks to my father, Suliman Alkhuraiji; my mother, Aljawharah Alkhalaf; my oldest brother, Abdullah Alkhuraiji; my wife, Munira Alkhuraiji; my cousin Abdulaziz Almalag; my family from the youngest to the oldest; and my friends who have supported and motivated me during my life journey up this moment. Without their love and support I would not be the person that I am today.

PAGE 6

vi TABLE OF CONTENTS CHAPTER I. INTRODUCTION ................................ ................................ ................................ ................................ ... 1 Overview ................................ ................................ ................................ ................................ ................. 1 Research Objectives ................................ ................................ ................................ ........................... 1 II. LITERATURE REVIEW ................................ ................................ ................................ ....................... 3 Introduction ................................ ................................ ................................ ................................ .......... 3 Bridge Deck ................................ ................................ ................................ ................................ ........... 3 Zinc Primer on Steel Decks ................................ ................................ ................................ ....... 3 III. EXPERIMENTAL WORK ................................ ................................ ................................ .................... 5 M ethods ................................ ................................ ................................ ................................ .................. 5 Samples Preparation ................................ ................................ ................................ .......................... 6 Steel PPC Panels Samples Preparation ................................ ................................ ................ 6 Steel Plates Preparation ................................ ................................ ................................ ... 7 Primer Preparation ................................ ................................ ................................ ............ 9 Polyester Polymer Concrete Preparation ................................ ............................... 10 Coring ................................ ................................ ................................ ................................ ... 12 Placing Pull off Discs ................................ ................................ ................................ ....... 16 Cylinder Samples Preparation ................................ ................................ ............................. 18 Steel Bond Samples Preparation ................................ ................................ ........................ 19 Samples Summary ................................ ................................ ................................ .................... 29 Machines Used for Testing ................................ ................................ ................................ .... 30 IV. TESTING ................................ ................................ ................................ ................................ ............... 32

PAGE 7

vii Steel Bond Test ................................ ................................ ................................ ................................ .. 32 Steel Bond Results ................................ ................................ ................................ .................... 34 Pull off Test ................................ ................................ ................................ ................................ ........ 36 Pull off Test Results ................................ ................................ ................................ ................. 38 Compression Test ................................ ................................ ................................ ............................. 55 Compression Test Results ................................ ................................ ................................ ..... 57 V. DISCUSSION ................................ ................................ ................................ ................................ ........ 64 Steel Bond Test Results ................................ ................................ ................................ ................. 64 Pull off Test Results ................................ ................................ ................................ ........................ 66 Compression Test Results ................................ ................................ ................................ ............. 70 Conclusion ................................ ................................ ................................ ................................ ........... 70 F uture Research ................................ ................................ ................................ ................................ 71 REFERENCES ................................ ................................ ................................ ................................ ....................... 72

PAGE 8

viii LIST OF TABLES TABLE 1 . The ultimate stress of the bond between the steel and the primer without zinc and the zinc primer. ................................ ................................ ................................ ................................ ...... 34 2 . The temperature that had been used for each steel PPC panel. ................................ ......... 37 3 . T he pull off test results in detail for each core of P1. ................................ .............................. 42 4 . T he pull off test results in detail for each core of P2. ................................ .............................. 43 5 . T he pull off test results in detail for each core of P3. ................................ .............................. 44 6 . T he pull off test results in detail for each core of P4. ................................ .............................. 45 7 . T he pull off test results in detail for each core of P5. ................................ .............................. 46 8 . T he pull off test results in detail for each core of P6. ................................ .............................. 47 9 . T he pull off test results in detail for each core of PZ1. ................................ ........................... 48 10 . T he pull off test results in detail for each core of PZ2. ................................ ........................... 49 11 . T he pull off test res ults in detail for each core of PZ3. ................................ ........................... 50 12 . T he pull off test results in detail for each core of PZ4. ................................ ........................... 51 13 . T he pull off test results in detail for each core of PZ5. ................................ ........................... 52 14 . T he pull off test results in detail for each core of PZ6. ................................ ........................... 53 15 . T he pull off test average of P1 to P6 . ................................ ................................ ............................. 54 16 . T he pull off test average of PZ1 to PZ6. ................................ ................................ ........................ 54 17 . T he total pull off test average of P1 to P6 and PZ1 to PZ6. ................................ ................... 54 18 . T he cylinder compression test results. ................................ ................................ ......................... 57

PAGE 9

ix LIST OF FIGURES FIGURE 1 . The final look of the steel PPC panels ready for P ull off test ................................ .................. 7 2 . The polished steel plate on the left and the original steel plate before preparation on the right. ................................ ................................ ................................ ................................ ....................... 8 3 . The wood formw ork preparation (a) shows the transparent tape covering the wood, (b) shows the formwork after assemblage ................................ ................................ .................... 8 4 . Applying the zinc primer into the steel plates. ................................ ................................ .......... 10 5 . The polyester resin and the catalyst preparation (a) shows while measuring the 3202 DDM9 before adding it to the polyester resin, (b) shows the polyester r esi n and the catalyst while it was mixing. ................................ ................................ ................................ ............. 11 6 . M ixing the components of the PPC. ................................ ................................ ................................ 12 7 . The twelve steel PPC panels after the PPC had been placed. ................................ ............... 12 8 . Each triangle corner simulates the Pull off machine legs location where the circles present s the discs locations. ................................ ................................ ................................ .............. 13 9 . The locations of each core on each steel PPC panel. ................................ ................................ 14 10 . Ten coring locations had been located on a steel PPC panel before the coring started. ................................ ................................ ................................ ................................ ................................ ...... 14 11 . T he coring process is going on. ................................ ................................ ................................ ........ 15 12 . Steel PPC panel after the cores had been drilled. ................................ ................................ ..... 16 13 . The bottom surface of an aluminum disc after it had been roughened. .......................... 17 14 . Two steel PPC panels and twenty discs just before being glued to each other by LOCTITE epoxy. ................................ ................................ ................................ ................................ ...... 18

PAGE 10

x 15 . ................................ ................................ ................. 19 16 . The first trial samples for steel bond test. ................................ ................................ ................... 20 17 . The steel bond samples from the second trial. ................................ ................................ .......... 21 18 . The third trial for the steel bond samples. ................................ ................................ .................. 22 19 . T he webs of the W sections had been cut from the middle to create WT sections. .... 23 20 . The steel bond samples that had been prepared from trial four. ................................ ...... 24 21 . C utting the flange to reduce the flange width from 4.06" to 1.85". ................................ .... 25 22 . The steel bond samples just before placing the primer between the flanges. .............. 26 23 . The steel temperature for the steel bond samples (a) shows the steel temperature just before placing the primer without zinc, (b) shows the steel temperature just before placing the zinc primer. ................................ ................................ ................................ ........ 28 24 . From the left; the 4 oz. of the primer, while the CHP had been adding to the primer, and 1/32 teaspoon of Z Cure. ................................ ................................ ................................ ........... 28 25 . The picture on the left shows the placing of the primer without zinc. The picture on the right shows the zinc primer just before it had been placed for the bond. ............... 29 26 . The picture on the left shows 6 SB samples for primer without zinc. The picture on the right shows 6 SBZ samples for the zinc primer. ................................ ................................ . 29 27 . The twelve steel PPC panels and the 120 cores with the pull off tester. ........................ 30 28 . The machines that were used for testing, (a) shows the 2 00 Kips MTS machine, (b) shows the 20 Kips MTS machin e, (c) shows DYNA Z16 pull off tester. ........................... 31 29 . One of the steel bond samples gripped by the MTS grippers and ready for the test. . 33 30 . The MTS controlling monitor after a sample had been tested, shown the displacement vs. time and the force vs. time graphs. ................................ .............................. 33

PAGE 11

xi 31 . The failure shape of each steel bond sample for the primer without zin c ..................... 35 32 . The failure shape of each steel bond sample for zinc primer ................................ .............. 36 33 . P2 and PZ2 while they heated up in the oven. ................................ ................................ ........... 38 34 . T esting one of the cores by DYNA Z16 pull off tester. ................................ ............................. 38 35 . A ll the cores data for the steel PPC panels bonded by primer without zinc. ................. 40 36 . A ll the cores data for the steel PPC panels bonded by zinc primer. ................................ .. 40 37 . A ll the core pull off results for the steel PPC panels bonded by prim er without zinc and zinc primer. ................................ ................................ ................................ ................................ ..... 41 38 . E ach pull are located the same as the location of clock numbers. ................................ ................................ ................................ ................. 42 39 . E ach pull are located the same as the location of clock numbers. ................................ ................................ ................................ ................. 43 40 . E ach pull are locat ed the same as the location of clock numbers. ................................ ................................ ................................ ................. 44 41 . E ach pull are located the same as the location of clock numbers. ................................ ................................ ................................ ................. 45 42 . E ach pull are located the same as the location of clock numbers. ................................ ................................ ................................ ................. 46 43 . E ach pull ers are located the same as the location of clock numbers. ................................ ................................ ................................ ................. 47 44 . E ach pull are located the same as the location of clock numbers. ................................ ................................ ................................ ................. 48

PAGE 12

xii 45 . E ach pull are located the same as the location of clock numbers. ................................ ................................ ................................ ................. 49 46 . E ach pull off failure on PZ3. Where are located the same as the location of clock numbers. ................................ ................................ ................................ ................. 50 47 . E ach pull are located the same as the location of clock numbers. ................................ ................................ ................................ ................. 51 48 . E ach pull are located the same as the location of clock numbers. ................................ ................................ ................................ ................. 52 49 . E ach pull off failu are located the same as the location of clock numbers. ................................ ................................ ................................ ................. 53 50 . T he laser extensometer that had been used for cylinder strain measurement. ........... 56 51 . Stress strain diagram for C1 ................................ ................................ ................................ .............. 58 52 . Stress strain diagram for C2 ................................ ................................ ................................ .............. 58 53 . Stress strain diagram fo r C3 ................................ ................................ ................................ .............. 59 54 . Stress strain diagram for C4 ................................ ................................ ................................ .............. 59 55 . Stress strain diagram for C5 ................................ ................................ ................................ .............. 60 56 . Stress strain diagram for C6 ................................ ................................ ................................ .............. 60 57 . The left picture showing C1 before the compression test starting and the picture on the right showing it after it failed. ................................ ................................ ................................ ... 61 58 . The left picture showing C2 before the compression test starting and the picture on the right showing it after it failed. ................................ ................................ ................................ ... 61 59 . The left picture showing C3 before t he compression test starting and the picture on the right showing it after it failed. ................................ ................................ ................................ ... 62

PAGE 13

xiii 60 . The left picture showing C4 before the compression test starting and the picture on the right showing it after it failed. ................................ ................................ ................................ ... 62 61 . The left picture showing C5 before the compression test starting and the picture on the right showing it after it failed. ................................ ................................ ................................ ... 63 62 . The left picture showing C6 before the compression test starting and the picture on the right showing it after it failed. ................................ ................................ ................................ ... 63 63 . ared with the steel bond average strength. ................................ ................................ ................................ ................................ ... 65 64 . T PPC panels bonded by primer without zinc and zinc primer. ................................ ................................ ................................ ........... 68 65 . ............................... 68 66 . T he total average of R 70,t/c for all cores results from the steel PPC panels bonded by pr imer without zinc and zinc primer. ................................ ................................ ............................ 69 67 . R 70,t/c for all cores results and its average for the steel PPC panels bonded by primer without zinc and zinc primer. Also, the theoretical ratio k ALK ................................ ............. 69

PAGE 14

1 CHAPTER I IN TRODUCTION Overview P olyester Polymer Concrete (PPC) which is made of mixing resin and aggregate components (United State Patent No. 4,816,503, 1989) has used PPC as an overlay on bridge decks . From the very beginning, damaged concrete used to be repaired by PPC, especially damaged concrete bridge decks (Alethafa, 2016) . PPC offers several advantages over other materials, i ncluding a high degree of water resistance, high compressive strength and workability. Used as an overlay for bridge decks, PPC protects the rebars in reinforced concrete decks and the steel plate s in steel bridge deck s. While PPC is a great material that can replace regular asphalt overlay s, it is too expensive to be used to replace all asphalt overlays. But it may be cost effective in many PPC can protect stee l decks from water, which is widely known to be an enemy of steel. T he direct cost of corrosion in the United State s is $278 billion. The cost of corrosion damage to infrastructure is $22.6 billion, and for highway bridges the cost is $8.3 billion . Where t he i ndirect costs to user s , such as traffic delays and lost productivity, have been estimated to be as high as 10 times that of direct corrosion costs (Koch, Brongers, Thompson, Virmani, and Payer, 2002) . In most cases, the environment must be controlled while applying PPC and the primer between the PPC and the bridge deck. This may cause problem s and minimize the uses of the PPC as overlay for concrete bridge decks. But for steel bridge decks , some

PAGE 15

2 preparations can be made off site, reducing closure times . The PPC used in this report n eeds only two hours to be hard before the bridge can be reopened to traffic. Research Objectives The overriding objective of this paper is to study the bond between a bridge steel deck and the PPC layer . T w o different bond primers were used -one without zinc and the other with zinc. Temperature was considered as major environmental exposure and its effect on the bond was tested . Additionally, analyzing t he p ure bond between the primer and the steel deck is one of the majo r objectives on this paper. Research involved twelve panels made of steel plate and PPC overlay. On six of the panels, the steel plates and PPC were bonded using primer without zinc and the rest were bonded using zinc primer. In addition, ten core samples were drilled on each panel for testing the s trength of the bonds between the steel and the PPC overlay. Six cylinders of PPC were made for testing compressive strength. These experiments lead to know ledge about the strength of bonds between steel plate s and primer s . In addition, the experiments clarify the effect s of temperature on PPC and on bond s between PPC and steel plate s .

PAGE 16

3 CHAPTER II LITERATURE REVIEW Introduction PPC has been used extensively overseas as a bridge deck overlay and in the U.S. on bo th concrete and steel bridge decks. Where some has been published about the physical properties of PPC. One exception was Alethafa (2016), which investigated PPC overlay provided by Kwik Bond Polymers on a concrete deck and steel deck, as well as the inter nal shear resistance, fatigue, and modulus of rupture of PPC. This study found that the average tensile strength of PPC was 696 psi, which was compressive strength was 4977 psi. The modulus of rupture was higher than the tensile strength, which was 50% of the PPC compressive strength. (The current study found the bond strength of the primer, which may signal an opportunity for further research.) Bridge Deck Aleth af a (2016) also addressed some specific properties related to different bridge types. Bridge decks are typically made of concrete, steel, wood or fibers. Concrete bridge decks are typically either cast in place or precast, while steel br idge decks are usually made of either orthotropic steel or steel plates. Alethafa (2016) explained that orthotropic steel decking is usually manufactured in a shop, where stiffeners are added to the steel plate base material. These stiffeners increase

PAGE 17

4 Aleth af a (2016) also noted that use of PPC as an overlay on steel decks provides protection against corrosion. This topic was explored in greater detail by another study. Z inc P rimer on St eel D ecks As is well known, steel is subject to corrosion, a problem that is compounded by the exposure of bridges to many environmental conditions, particularly humidity and water . According to Albrecht and Hall ( 2003 ) using zinc primer as a base coat pr otects steel decks from water exposure and is required by most state highway departments.

PAGE 18

5 CHAPTER III EXPERIMENTAL WORK Methods The experimental work for this analysis focused on the bond strength between steel plates and primers with and without zinc. In addition, experiments addressed the thermal effects on the bond between the steel plates and PPC, as well as the compressive strength of PPC. The steel plates for the panels were polished by grinder metal brush to simulate the surface of steel bridge decks and then PPC measuring 2 inches thick was placed on steel plates. Then ten core samples were drilled on each of the twelve steel plates using the following method. The coring was done by a using a 2 inch diameter coring bit, which made core samples that were 1.767 inche diameter. Each of the cores was drilled all the way through the PPC until the bit touched the steel plate. These 120 core samples were used in pull off tests. Samples of steel PPC panels that were bonded by primer with zinc and other samples that were bonded by primer without zinc were tested at six different temperatures tested. Ten samples for each test were from a panel with the zinc primer and ten were from a panel with the zinc free primer. In addition, six cylinders of PPC were created to test the compressive strength of the material.

PAGE 19

6 Samples Preparation Steel PPC Panels Samples Preparation The steel PPC panels were made of steel plates measu ring 12 by 12 by 3/8 inches and PPC overlay that was 2 inches thick. A total of twelve steel PPC panels was made. Six of them were bonded by primer without zinc and the other six were bonded by zinc primer. Then, ten cores were drilled on each steel PPC pa nel. Each core has a diameter of 1.767 inches. The location of each core was chosen carefully to make sure the pull off tester could be placed on the steel PPC panel (See Figure 1 ) Each steel PPC panel was labeled 86, PZ4 112). The panel zinc ed by the primer

PAGE 20

7 Figure 1 . The final look of the steel PPC panels ready for Pull off test Steel Plate Preparation The surface of the twelve steel plates was prepared to simulate the surface of a steel bridge deck. A metal brush was used to polish the rust and the thin layer that covered the hot rolled plate, as seen Figure 2 . This was done to make sure the bond between the ste el plate and the PPC was a good simulation of a steel bridge deck bond. After the steel plates were ready, formwork preparation was required.

PAGE 21

8 Figure 2 . The polished steel plate on the left and the original steel plate before prep aration on the right. Formwork preparation. Before placing the PPC overlays on the steel plates, a wood formwork was prepared for each steel PPC panel. The inside face of the formwork was covered by transparent tape to prevent any bond between the PPC and the wood ( Figure 3 ). After preparing the formwork, the primers and the PPC could be prepared. (a) (b) Figure 3 . The wood formwork preparation (a) shows the transparent tape covering the wood, (b) shows the formwork after assemblage

PAGE 22

9 Primer Preparation Two primers were used in this experiment one with and the other without zinc. The primer components were measured by volume unless the zinc was measured by weight. The ma terial was provided by Kwik Bond Polymers (KBP), Inc. The temperature of Preparation of the primer without zinc began with combining 12 ounces of KBP Pro Prime 18 905, whic teaspoon of Z Cure, which was the accelerator. These were mixed together well. After mixing the primer without zinc, a paint brush was used to apply the primer without zinc to six steel plates. After that, 4.5 pounds of zinc were added to the mix of the primer without zinc to make the zinc primer. Then the zinc primer was applied to the remaining six steel plates ( Figure 4 .)

PAGE 23

10 Figure 4 . Applying the zinc primer i nt o the steel plates . Polyester Polymer Concrete Preparation After placing the primer without zinc and the zinc primer to the steel plates, the PPC was placed on the steel plates to make the steel PPC panels. The preparation for the PPC was around 11a.m. The PPC that had been used on this experimental work was made of 150 pounds of aggregates; two gallons of polyester resin, which was the binder; and five ounces of 3202 DDM9, which was the catalyst. As Figure 5 shows, the polyester resin and the catalyst were combined first, and then the mix was added to the aggregates. All the components were mixed very well, as seen in Figure 6 . Th e PPC overlay thickness was 2

PAGE 24

11 inches. After placement of the PPC, the steel PPC panels were left in place until the next morning, as indicated in Figure 7 . At that time, the formwork was removed. (a) (b) Figure 5 . The polyester resin and the catalyst preparation (a) shows w hile measuring the 3202 DDM9 before adding it to the polyester resin, (b) shows the polyester r e s i n and the catalyst while it was mix ing .

PAGE 25

12 Figure 6 . M ixing the components of the PPC. Figure 7 . The twelve steel PPC panels after the PPC had been placed. Coring The coring was drilled on the steel PPC panels using a 2 inch diameter coring bit to create 1.767 i nch diamter core samples. Each core went all the way through until the

PAGE 26

13 coring bit touched the steel surface. There were ten cores on each steel PPC panel. Their locations were chosen carefully to make sure the pull off machine could be placed in a position that would help yield reliable test results ( Figure 8 and Figure 9 ). After the location of each core on a steel PPC panel had been designated, the coring was begun ( Figure 10 and Figure 11 ). Finally, the steel PPC panels were ready for placement of the discs for pull off testing, as indicated in Figure 12 . Figure 8 . Each triangle corner simulates the Pull off machine legs location w here the circles present s the discs locations.

PAGE 27

14 Figure 9 . The locations of each core on each steel PPC panel. Figure 10 . Ten coring locations had been located on a steel PPC panel before the coring started.

PAGE 28

15 Figure 11 . T he coring process is going on.

PAGE 29

16 Figure 12 . S teel PPC panel after the cores h ad been d rilled . Placing Pull off Discs For the pull off test, 2 inch diameter aluminum discs were placed on each core, as in Figure 1 . The total number of the discs that had been used on this experiment was 120. The bottom surface of each disc had been roughened to be sure the bond between the discs and the PPC cores was strong enough, as indicated in Figure 13 . The strong bond between the discs and the PPC forced the fail ure to happen on the steel PPC bond surface or on the PPC itself. The discs were attached to the PPC cores by using LOCTITE heavy duty epoxy, which has 3500 psi strength, as indicated in Figure 14 . After the discs were placed on the steel PPC panels, the panels were ready for the pull off test. The discs had been labeled by ten locations on watches, as in Figure 1 . The final information on each disc included the panel

PAGE 30

17 number, including temperature, followed by the disc number (Example: P3 102 5, PZ2 86 11). Figure 13 . The bottom surface of an aluminum disc after it ha d b een roughened.

PAGE 31

18 Figure 14 . Two steel PPC panels and twenty discs just before being glued to each other by LOCTITE epoxy. Cylinder Samples Preparation While preparing the steel PPC panels, six cylinders of PPC were prepared as sho wn in Figure 15 . The cylinders were 4 inches in diameter and 8 inches deep. A tamping rod was used while making the cylinders to prevent any voids. When the PPC cylinders had hardened, their circular surfaces were leveled by using a saw to make sure the stress on the cylinders would be uniform. Then the cylinders were left for 328 days before the me from the initial of the word cylinder .

PAGE 32

19 Figure 15 . After preparing six Steel Bond Samples Preparation The steel bond sample preparation was the biggest challenge in the experimental work. The steel bond samples were designed to test the pure bond between the steel deck and both the primer without zinc and the zinc primer. The issue was how to simulate the steel deck surface? In addition, what is the best way to assemble the samples? Many ideas were tried to make these samples. For the first trial, a steel plate was prepared in the same way as the one in Figure 2 . It was then cut into four 6 inch by 6 inch pieces. The primers were placed on these steel plates. The n stainless steel discs were placed on the primers, as indicated in Figure 16 . After the samples had hardened, three holes were drilled in each steel plate to make it fit in the Mechanical Testing and Sensing Solut ions ( MTS ) machine and to make sure the disc would be exactly in the middle. These samples did not show any results of primer bond strength to the steel . The bottom surface of the stainless steel discs was very smooth and did not simulate the steel deck su rface .

PAGE 33

20 Figure 16 . The first trial samples for steel bond test . The second trial followed a similar process, replacing the stainless steel discs with 2 inch by 2 inch plates. These plates had welded steel nuts with threads for bo lts to apply tension force on the bond. The bottom surface of the 2 inch by 2 inch steel plates was leveled to make sure the welded nut was perpendicular to bond surface. This was done to prevent any bending moment on the bond. In addition, some silicone w as added on the edges of the 6 inch by 6 inch steel plates to control the primer and prevent it from spilling off of the plate, as indicated in Figure 17 . Unfortunately, all results from this trial failed, too. In other word, primer bond strength to the steel could not be evaluated. The bottom surface of the 2 inch by2 inch steel plate was polished and that made the surface too smooth.

PAGE 34

21 Figure 17 . The steel bond samples from the second tria l . For the third trial, five samples were prepared using two stainless steel discs for each sample. The bottom surface of each disc was roughened by grinder to increase the bond between the primer and the discs. Then one tenth to one eighth inch of primer was placed between the two discs to create a single sample, as shown in Figure 18 . The bond was placed on 2 inch diameter discs, each with a roughened surface area of 3.14 square inches. After the material hardene d, the test was run and the results were impressive , better than the previous trials . Then another five samples were prepared with the same primer. With the second five samples, testing showed a 31% increase in stress over the first five samples -around 3 91 psi more than the first five samples . Thus, the bottom surface roughness in the first five samples was not the same as the roughness on the second five samples. The roughness was randomly created and did not simulate the steel deck surface, resulting in an inaccurate trial. The average stress from this trial was 1454 psi, and failure typically appeared to occur in the material itself, not in the bond .

PAGE 35

22 Figure 18 . The third trial for the steel bond samples. For the fourth trial, a small I beam (W4x13) was sandblasted and cut into twenty middle to create two tee sections, as in Fi gure 19 . The sandblasted f lange outer face surface of each tee section was used to simulate the steel deck surface. Between the flanges of two tee sections, one tenth to one eighth inches of primer had been placed, as in Figure 20 . The cros s sectional area of the bond was 1½ inches by 4.06 inches with an area of 6.09 square inches. The test was run two days after the samples had been prepared. The results were incredible and did not make any sense. The failure was typically on the primer it self, and the average stress was 608.8 psi. This stress measurement was 58% (845 psi) lower than that in trial three, where the samples had almost half as much bond area. The reason behind this significant decreasing in the stress was the flange rigidity. The flange was flexible compared to the stainless steel discs. In other words, the cross sectional

PAGE 36

23 area of the bond was not 100 percent active because of the stress distribution on the bond surface. This trial was the best and the easiest to maintain . Fi gure 19 . T he web s of the W section s had been cut from the middle to create WT sections.

PAGE 37

24 Figure 20 . The steel bond samples that had been prepared from trial four . For the fifth and final trial, the fourth trial was repeated but with shorter flange width, as shown in Figure 21 . The effective flange width had been calculated based on the average applied force from trial four (3,707.3 lbf) and the average stress from third trial (1,454 psi). The length of each tee section flange was 1½ inches. In other words, trial five was designed to determine how the same stress value as trial three could be achieved using the same preparation as trial four. The effective flange wid th was calculated as follows:

PAGE 38

25 It was clea r the sandblasted surface roughness would never have higher roughness than the discs on trial three. So, the effective flange width had been taken as 1.85 inches, which should be reasonable. This flange width had made a bond surface area of 2.78 square inc hes. Figure 21 . C utting the flange to reduce the flange width from 4.06" to 1.85". After the shape and the bond surface area had been determined, the way to lenge. Each At the same time, the webs of the two tees of each sample had to be aligned to prevent any

PAGE 39

26 bending moment on the bond surface. Also, the webs needed to align correctly for placement on the MTS machine for testing. section. The length of each was perpendicular to the Figure 21 . inch thick. Also, the outer face of each flange was cleaned very well by using Acetone. This helped to create the best bond between the steel and the prim er. Then, a form for each sample was made by using a tape, as in Figure 22 . Figure 22 . The steel bond samples just before placing the primer between the flanges. After the geometry and t he form of the samples were prepared, it was time to place the primer between the flanges to create the bond. The steel temperature needed to be between 6

PAGE 40

27 Figure 23 a. For the samples with the zinc Figure 23 b. For the primer without zinc. a KBP specialist advised combining 16 ounces of Pro Cure. Guidance called for the zinc primer to be made by adding 6.125 pound s of zinc to the same components and quantities as the primer without zinc. These quantities were too much for the samples that had been made for the bond test. The quantities were revised to be 4 ounces of Pro Prime, 9/16 teaspoon of CHP, and 1/32 teaspoo n of Z Cure for the primer without zinc, as in Figure 24 . The same revised quantities were used for the primer with zinc and 1.531 pounds of zinc were added. The primer was applied to fill in the gaps between the t harden, as in Figure 25 . Each bond sample was labeled by SB followed by a number (Example: SB3). SB came from the initials of steel and bond . The steel bond samples with zinc primer were the initial from word zinc . By then, the steel bond samples had been prepared and they were ready for testing, as in Figure 26 .

PAGE 41

28 (a) (b) Figure 23 . The steel temperature for the steel bond samples (a) shows the steel temperature just before placing the primer without zinc , (b) shows the steel temperature just before placing the zinc primer . Figure 24 . From the left; the 4 oz. of the primer, while the CHP had been adding to the primer, and 1/32 teaspoon of Z Cure.

PAGE 42

29 Figure 25 . The picture on the left shows the placing of the primer without zinc. The pic ture on the right shows the zinc primer just before it had been placed for the bond. Figure 26 . The picture on the left shows 6 SB samples for primer without zinc. The picture on the right shows 6 SB Z samples for the z inc primer. Samples Summary The samples had been prepared for this experimental work are summarized in the list below: 1 . Six steel PPC panels bonded with primer without zinc. Each panel had 10 cores of 1.7 7 . Which ma ke it 60 cores in total , as in Figure 27 .

PAGE 43

30 2 . Six steel PPC panels bonded by zinc primer. Each panel had 10 cores of 1.7 7 . Which ma k e it 60 cores in total , as in Figure 27 . 3 . Six PPC cylinders. 4 . Six steel bond samples bonded with primer without zinc , as in Figure 26 . 5 . Six steel bond samples bonded with zinc primer , as in Figure 26 . Figure 27 . The twelve steel PPC panels and the 120 cores with the pull off tester. Machines Used for Testing Three testing machines were used in this experimental work. The 2 00 Kips MTS machine was used for PPC cylinder tests , as in Figure 28 a . The 20 Kips MTS machine was used for the steel bond sample tests , as in Figure 28 b . The DYNA Z16 p ull off test er that was used to test the discs on the steel PPC panels , as in Figure 28 c .

PAGE 44

31 (a) (b) (c) Figure 28 . The machines that were used for testing, (a) shows the 2 00 Kips MTS machine, (b) shows the 20 Kips MTS machine , (c) shows DYNA Z16 pull off tester.

PAGE 45

32 CHAPTER IV TESTING Steel Bond Test The SB samples were prepared and ready for testing, using the 20 Kips MTS machine, as Figure 29 shows. As noted in Chapter III, special care was taken to align the webs of the SB samples because the upper and the lower grippers of the MTS machine are rigid and perfectly aligned. Before testing could begin, a procedure had to be established for applying force on the samples. This force could be applied by force per time or displacement per time. For this experimental work, a force rate of 10 pounds per second was used with the expectation that this force rate would assure that any failure wou ld be a bond failure. After the procedure had been set up and the MTS grips were attached, the bond test was ready to be run, as reflected in Figure 29 . When the test started running, displacement vs. time and forc e vs. time diagrams appeared on the MTS computer, as seen in Figure 30 .

PAGE 46

33 Figure 29 . One of the steel bond samples gri p ped by the MTS grips and ready for the test. Figure 30 . The MTS controlling monitor after a sample had been tested, shown the displacement vs. time and the force vs. time graphs 1 . 1 The results on this picture were for a steel bond sample from trail 3.

PAGE 47

34 Steel Bond Results There were 12 steel bond samples -six of them bonded by primer without zinc and six bonded with z inc primer. The results are shown in Table 1 . Table 1 . The ultimate stress of the bond between the steel and the primer without zinc and the zinc primer . Primer Without Zinc Zinc Primer Ste el Bond Sample Ultimate Force (lbf) Ultimate Bond Stress (psi) Steel Bond Sample Ultimate Force (lbf) Ultimate Bond Stress (psi) SB1 3,397.7 1,222.2 SBZ1 3,603.9 1,296.4 SB2 2,697.9 970.5 SBZ2 3,045.5 1,095.5 SB3 2,769.2 996.1 SBZ3 3,952.5 1,421.8 SB 4 2,996.5 1,077.9 SBZ4 4,472.5 1,608.8 SB5 3,015.1 1,084.6 SBZ5 3,784.6 1,361.4 SB6 2,964.8 1,066.5 SBZ6 4,305.2 1,548.6 Average 2,973.5 1,069.6 Average 3,860.7 1,388.7 The ultimate bond stress had been calculated base d on 2.78 in 2 bond area. The fa ilure shapes of the steel bond samples that were bonded with primer without zinc are shown on Figure 31 , while Figure 32 shows the failure shape for the samples bonded with zinc primer.

PAGE 48

35 (a) SB1 (b) SB2 (c) SB3 (d) SB4 (e) SB5 (f) SB6 Figure 31 . The failure shape of each steel bond sample for the primer without zinc

PAGE 49

36 (a) SB Z 1 (b) SB Z 2 (c) SB Z 3 (d) SB Z 4 (e) SB Z 5 (f) SB Z 6 Figure 32 . The failure shape of each steel bond sample for zinc p rimer Pull off Test The DYNA Z16 Pull off tester was used to determine the bond strength be tween the steel and the PPC on the steel PPC panels. The pull off tester was adjusted to present the force in force pound units because the area of each core was less than 3.14 square inches 1 . 1 The DYNA Z16 Pull re, the machine was adjusted to present force only in lbf units because the cores in this experimental work had

PAGE 50

37 After the aluminum discs had been glued onto the cores in the s teel PPC panels, it was the Table 2 . Table 2 . The temperature that had been used for each steel PPC panel. Primer Without Zinc Zinc Primer Panel Average Panel Average P1 70 PZ1 70 P2 86 PZ2 86 P3 102 PZ3 102 P4 112 PZ4 112 P5 120 PZ5 120 P6 160 PZ6 154 rest of the samples were placed in an oven, two panels at a time, as Figure 33 shows. Each panel did not have a constant temperature. The temperature may have varied +/ because each core required time to be tested and the temperature decreased during this time. Typically, five cores from each panel were tested in sequence, and then the panel was returned to the oven to control its temperature. For example, P3 was taken out of the oven for testing and five cores were teste d by the pull off tester. P3 was then returned to the oven and PZ3 was taken from the oven for testing five cores of it. Then PZ3 was returned to the oven and P3 was taken out of the oven for testing of its five remaining cores. Finally, PZ3 was removed fr om the oven for testing of its five remaining cores. As seen in Figure 34 , while the pull off test was going.

PAGE 51

38 Figure 33 . P 2 and PZ 2 while they heated up in the oven. Figure 34 . T esting one of the cores by DYNA Z16 pull off tester. Pull off Test Results When the 120 sample pull off test were completed, as indicated in Figure 27 , eleven of the cores failed at the int erface between the epoxy and the disc. Some of the failed disc results were dismissed, while others were not. Results were not dismissed where the

PAGE 52

39 failure load was close to the other samples and it was under estimation to the PPC tensile strength in a spec ific steel PPC panel with reasonable amount. temperature of the steel PPC panel. Then the temperature decreased to the average temperature at the fifth core. After the fifth core w as tested, the panel was reheated in the oven until the sixth core reached roughly the same temperature as the fifth core had been. PPC panel to the tenth core. The temperature varia tion was observed on all steel PPC panels excluding P1 and PZ1, which were tested at room temperature. Results for the steel PPC panels bonded with the primer without zinc are summarized in Figure 35 , while detail ed pull off results for P1 through P6 are reflected in Table 3 through Table 8 and Figure 38 through Figure 43 . Results for the steel PPC panels bonded with zinc primer are summarized in Figure 36 , while detailed pull off results for PZ1 through PZ6 are reflected in Table 9 through Table 14 and Figure 44 through Figure 49 . All the pull off test results from P1 through P6 and PZ1 through PZ6 were merged and compared with the temperature of the steel PPC panels in Figure 37 . The stress was calculated based on a 2.45 in 2 core cross sectional area. Table 15 sh ows the pull off test average for the panels bonded with primer without zinc, while Table 16 shows the pull off test average for the zinc primer panels. The average based on temperature values for all the pull off test results is shown in Table 17 .

PAGE 53

40 Figure 35 . A ll the cores data for the steel PPC panels bonded by primer without zinc. Figure 36 . A ll the cores data for the s teel PPC panels bonded by zinc primer. 0 100 200 300 400 500 600 700 800 900 60 70 80 90 100 110 120 130 140 150 160 170 Tensile Strength, psi Pull Off Test Results for Primer Without Zinc Panels Test Results Results Average 0 100 200 300 400 500 600 700 800 900 1000 60 70 80 90 100 110 120 130 140 150 160 170 Tensile Strength, psi Pull Off Test Results for Zinc Primer Panels Test Results Results Average

PAGE 54

41 Figure 37 . A ll the core pull off results for the steel PPC panels bonded by primer without zinc and zinc primer. 0 100 200 300 400 500 600 700 800 900 1000 60 70 80 90 100 110 120 130 140 150 160 170 Tensile Strength, psi All Pull Off Test Results (Primer Without Zinc & Zinc Primer) Primer Without Zinc Results Zinc Primer Results

PAGE 55

42 Table 3 . T he pull off test results in detail for ea ch core of P1. Disc Number Temperature Load (lbf) Time (sec.) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode 1 70 1,983.0 263.0 7.5 809.4 PPC 2 70 1,737.0 249.0 7.0 709.0 PPC + P 3 70 1,614.0 210.0 7.7 658.8 PPC 4 70 1,603.0 250.0 6.4 6 54.3 PPC 5 70 1,898.0 247.0 7.7 774.7 PPC + P 7 70 1,620.0 225.0 7.2 661.2 PPC + P 8 70 1,848.0 251.0 7.4 754.3 PPC + P 9 70 1,731.0 255.0 6.8 706.5 PPC 10 70 1,924.0 190.0 10.1 785.3 PPC 11 70 2,065.0 311.0 6.6 842.9 PPC Average 70 1 , 802.3 245.1 7. 4 735.6 80% PPC Number 6 is missing on purpose, because the discs numbers had been placed based on the numbers locations on a watch. PPC : means the fa ilure in the Polyester Polymer Concrete itself. PPC+P: means the failure happened between the PPC an d the Primer. Figure 38 . E ach pull off failure on P1. Where the discs numbers are located the same as the location of clock numbers.

PAGE 56

43 Table 4 . T he pull off test results in detail for each core of P 2. Disc Number Temperature Load (lbf) Time (sec.) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode 1 90 1 , 772.0 150.0 11.8 723.3 PPC 2 89 1 , 860.0 184.0 10.1 759.2 PPC 3 88 1 , 644.0 123.0 13.4 671.0 PPC 4 87 1 , 819.0 163.0 11.2 742.4 PPC + P 5 86 E 7 86 1 , 720.0 200.0 12.9 702.0 PPC + P 8 85 1 , 462.0 133.0 11.0 596.7 PPC 9 84 1 , 544.0 141.0 11.0 630.2 PPC + P 10 83 1 , 772.0 140.0 12.7 723.3 E 11 82 1 , 854.0 161.0 11.5 756.7 PPC Average 86 1 , 716.3 155.0 11.7 700.5 83 % PPC Number 6 is missing o n purpose, because the discs numbers had been placed based on the numbers locations on a watch. PPC: means the failure in the Polyester Polymer Concrete itself. PPC+P: means the failure happened between the PPC and the Primer. E: means the failure happe ned between the disc and the epoxy. Disc number 5 results had been dismissed in this paper because the epoxy failed . Where the results from disc number 10 had been used in this paper because its stress within range of the other discs stress es . The average had been calculated based on 9 s amples out of 10. Figure 39 . E ach pull off failure on P2. Where the discs numbers are located the same as the location of clock numbers.

PAGE 57

44 Table 5 . T he pull off test results in detail for each core of P 3. D isc Number Temperature Load (lbf) Time (sec.) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode 1 106 819.0 91.0 9.0 334.3 PPC 2 105 953.0 122.0 7.8 389.0 PPC 3 104 871.0 116.0 7.5 355.5 PPC 4 103 1 , 053.0 104.0 10.1 429.8 PPC 5 102 1 , 076.0 77.0 14.0 439.2 PPC 7 102 994.0 99.0 10.0 405.7 PPC + P 8 101 1 , 088.0 114.0 9.5 444.1 PPC 9 100 1 , 024.0 70.0 14.6 418.0 PPC 10 99 1 , 018.0 87.0 11.7 415.5 PPC 11 98 994.0 98.0 10.1 405.7 PPC + P Average 102 989.0 97.8 10.4 403.7 90% PPC Number 6 is missing o n purpose, because the discs numbers had been placed based on the numbers locations on a watch. PPC: means the failure in the Polyester Polymer Concrete itself. PPC+P: means the failure happened between the PPC and the Primer. Figure 40 . E ach pull off failure on P3. Where the discs numbers are located the same as the location of clock numbers.

PAGE 58

45 Table 6 . T he pull off test results in detail for each core of P 4. Disc Number Temperature Load (lbf) Time (sec.) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode 1 116 725.0 111.0 6.5 295.9 PPC 2 115 795.0 51.0 15.6 324.5 PPC 3 114 760.0 100.0 7.6 310.2 PPC 4 113 801.0 60.0 13.4 326.9 PPC 5 112 766.0 90.0 8.5 312.7 PPC 7 112 743.0 89.0 8.3 303. 3 PPC 8 111 731.0 89.0 8.2 298.4 PPC 9 110 760.0 89.0 8.5 310.2 PPC 10 109 819.0 91.0 9.0 334.3 PPC 11 108 819.0 75.0 10.9 334.3 PPC Average 112 771.9 84.5 9.7 315.1 100% PPC Number 6 is missing on purpose, because the discs numbers had been place d based on the numbers locations on a watch. PPC: means the failure in the Polyester Polymer Concrete itself. Disc number 4: the failure happened in the PPC. Then suddenly while trying to take off the core the epoxy had been failed that why this core result had been used in this paper. Figure 41 . E ach pull off failure on P4. Where the discs numbers are located the same as the location of clock numbers.

PAGE 59

46 Table 7 . T he pull off test results in detail for each core of P 5. Disc Number Temperature Load (lbf) Time (sec.) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode 1 124 603.0 54.0 11.2 246.1 PPC 2 123 643.0 86.0 7.5 262.4 PPC 3 122 632.0 102.0 6.2 258.0 PPC 4 121 643.0 100.0 6.4 262.4 PPC 5 120 661.0 63.0 10.5 269.8 PPC 7 120 638.0 102.0 6.3 260.4 PPC 8 119 655.0 75.0 8.7 267.3 PPC 9 118 643.0 95.0 6.8 262.4 PPC 10 117 678.0 91.0 7.5 276.7 PPC 11 116 673.0 96.0 7.0 274.7 PPC Average 120 646.9 86.4 7.8 264.0 100% PPC Number 6 is missing on purpose, because the discs numbers had been placed based on the numbers locations on a watch. PPC: means the failure in the Polyester Polymer Concrete itself. Figure 42 . E ach pull off failure on P5. Where the discs numbers are located the same as the location of clock numbers.

PAGE 60

47 Table 8 . T he pull off tes t results in detail for each core of P 6. Disc Number Temperature Load (lbf) Time (sec.) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode 1 164 140.0 38.0 3.7 57.1 PPC 2 163 164.0 37.0 4.4 66.9 PPC 3 162 158.0 32.0 4.9 64.5 PPC 4 161 164.0 38.0 4.3 66.9 PPC 5 160 170.0 42.0 4.0 69.4 PPC 7 1 60 170.0 41.0 4.1 69.4 PPC 8 159 158.0 42.0 3.8 64.5 PPC 9 158 158.0 43.0 3.7 64.5 PPC 10 157 164.0 45.0 3.6 66.9 PPC 11 156 170.0 46.0 3.7 69.4 PPC Average 1 6 0 161.6 40.4 4.0 66.0 100% PPC Number 6 is missing on purpose, because the discs numbers had been placed based on the numbers locations on a watch. PPC: means the failure in the Polyester Polymer Concrete itself. Figure 43 . E ach pull off failure on P6. Where the discs numbers are located the same as the location of clock numbers.

PAGE 61

48 Table 9 . T he pull off test results in detail for each core of P Z1. Disc Number Temperature Load (lbf) Time (sec.) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode 1 70 2 , 082.0 272.0 7.7 849.8 PPC 2 70 1 , 930.0 244.0 7.9 787.8 PPC 3 70 2 , 018.0 271.0 7.4 823.7 PPC 4 70 2 , 211.0 326.0 6.8 902.4 PPC 5 70 2 , 094.0 295.0 7.1 854.7 PPC 7 70 2 , 018.0 306.0 6.6 823.7 PPC 8 70 2 , 076.0 264.0 7.9 847.3 PPC 9 70 2 , 129.0 319.0 6.7 869.0 PPC 10 70 2 , 234.0 268.0 8.3 911.8 PPC 11 70 2 , 018.0 210.0 9.6 823.7 PPC Average 70 2 , 081.0 277.5 7.6 849.4 100% PPC Number 6 is missing on purpose, because the discs numbers had been placed based on the numbers locations on a watch. PPC: means the failure in the Polyester Polymer Concrete itself. Figure 44 . E ach pull off failure on PZ1. Where the discs numbers are located the same as the location of clock numbers.

PAGE 62

49 Table 10 . T he pull off test results in detail for each co re of P Z2. Disc Number Temperature Load (lbf) Time (sec.) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode 1 90 1597.0 651.8 E 2 89 1474.0 144.0 10.2 601.6 PPC 3 88 1901.0 775.9 E 4 87 1497.0 611.0 E 5 86 1579.0 160.0 9.9 644.5 PPC 7 86 1696.0 169.0 10.0 692.2 E 8 85 1521.0 117.0 13.0 620.8 E 9 84 1451.0 148.0 9.8 592.2 PPC + P 10 83 1708.0 161.0 10.6 697.1 PPC 11 82 1755.0 120.0 14.6 716.3 E Average 86 1617.9 145.6 11.2 660.4 95% PPC Number 6 is missing on purpose, because the discs numbers h ad been placed based on the numbers locations on a watch. PPC: means the failure in the Polyester Polymer Concrete itself. PPC+P: means the failure happened between the PPC and the Primer. E: means the failure happened between the disc and the epoxy. T he results from disc number 1, 3, 4, 7, 8, 10 and 11 had been used in this paper because its stress within range of the other discs stress es and its results conservative for the PPC tensile strength . The time and the force rate average had been calculate d based on 7 samples out of 10. Where the rest based on 10 samples. And the epoxy failures had been assumed as a PPC failure conservative . Figure 45 . E ach pull off failure on PZ2. Where the discs numbers are located the same a s the location of clock numbers.

PAGE 63

50 Table 11 . T he pull off test results in detail for each core of P Z3. Disc Number Temperature Load (lbf) Time (sec.) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode 1 106 907.0 122.0 7.4 370.2 PPC 2 105 784.0 105.0 7.5 320.0 PPC 3 104 743.0 103.0 7.2 303.3 PPC 4 103 1 , 018.0 109.0 9.3 415.5 PPC 5 102 1 , 070.0 128.0 8.4 436.7 PPC 7 102 936.0 97.0 9.6 382.0 PPC + P 8 101 889.0 93.0 9.6 362.9 PPC 9 100 1 , 064.0 124.0 8.6 434.3 PPC 10 99 1 , 094.0 132.0 8.3 446.5 PPC 11 98 1 , 029.0 121.0 8.5 420.0 PPC Average 102 953.4 113.4 8.4 389.1 95% PPC Number 6 is missing on purp ose, because the discs numbers had been placed based on the numbers locations on a watch. PPC: means the failure in the Polyester Polymer Concrete itself. PPC+P: means the failure happened between the PPC and the Primer. Figure 46 . E ach pull off failure on PZ3. Where the discs numbers are located the same as the location of clock numbers.

PAGE 64

51 Table 12 . T he pull off test results in detail for each core of P Z4. Disc Number Temperature Load (lbf) Time (s ec.) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode 1 116 702.0 87.0 8.1 286.5 PPC 2 115 655.0 88.0 7.4 267.3 PPC 3 114 673.0 86.0 7.8 274.7 PPC 4 113 690.0 90.0 7.7 281.6 PPC 5 112 696.0 86.0 8.1 284.1 PPC 7 112 673.0 87.0 7.7 274.7 PPC 8 111 667.0 88.0 7.6 272.2 PPC 9 110 702.0 96.0 7.3 286.5 PPC 10 109 696.0 92.0 7.6 284.1 PPC 11 108 684.0 88.0 7.8 279.2 PPC Average 112 683.8 88.8 7.7 279.1 100% PPC Number 6 is missing on purpose, because the discs numbers had been placed based o n the numbers locations on a watch. PPC: means the failure in the Polyester Polymer Concrete itself. Figure 47 . E ach pull off failure on PZ4. Where the discs numbers are located the same as the location of clock numbers.

PAGE 65

52 Table 13 . T he pull off test results in detail for each core of P Z5. Disc Number Temperature Load (lbf) Time (sec.) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode 1 124 608.0 67.0 9.1 248.2 PPC 2 123 643.0 72.0 8.9 262.4 PPC 3 122 584.0 75.0 7.8 238.4 PPC 4 121 602.0 55.0 10.9 245.7 PPC 5 120 572.0 61.0 9.4 233.5 PP C 7 120 619.0 69.0 9.0 252.7 PPC 8 119 608.0 101.0 6.0 248.2 PPC 9 118 620.0 101.0 6.1 253.1 PPC 10 117 649.0 98.0 6.6 264.9 PPC 11 116 643.0 108.0 6.0 262.4 PPC Average 120 614.8 80.7 8.0 250.9 100% PPC Number 6 is missing on purpose, because the discs numbers had been placed based on the numbers locations on a watch. PPC: means the failure in the Polyester Polymer Concrete itself. Figure 48 . E ach pull off failure on PZ5. Where the discs numbers are located the same as the location of clock numbers.

PAGE 66

53 Table 14 . T he pull off test results in detail for each core of P Z6. Disc Number Temperature Load (lbf) Time (sec.) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode 1 158 175.0 30.0 5. 8 71.4 PPC 2 157 187.0 46.0 4.1 76.3 PPC 3 156 170.0 44.0 3.9 69.4 PPC 4 155 187.0 45.0 4.2 76.3 PPC 5 154 187.0 44.0 4.3 76.3 PPC 7 154 181.0 42.0 4.3 73.9 PPC 8 153 193.0 48.0 4.0 78.8 PPC 9 152 193.0 44.0 4.4 78.8 PPC 10 151 211.0 63.0 3.3 86.1 PPC 11 150 199.0 59.0 3.4 81.2 PPC Average 154 188.3 46.5 4.2 76.9 100% PPC Number 6 is missing on purpose, because the discs numbers had been placed based on the numbers locations on a watch. PPC: means the failure in the Polyester Polymer Concret e itself. Figure 49 . E ach pull off failure on PZ6. Where the discs numbers are located the same as the location of clock numbers.

PAGE 67

54 Table 15 . T he pull off test average of P 1 to P6. Steel PPC Panel Temperatu re Load (lbf) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode P 1 70 1 , 802.3 7.4 735.6 80% PPC P 2 86 1 , 716.3 11.7 700.5 83% PPC P 3 102 989.0 10.4 403.7 90% PPC P 4 112 771.9 9.7 315.1 100% PPC P 5 120 646.9 7.8 264.0 100% PPC P6 160 161.6 4.0 66.0 100% PPC PPC: means the failure in the Polyester Polymer Concrete itself. Table 16 . T he pull off test average of P Z1 to PZ6. Steel PPC Panel Temperature Load (lbf) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode PZ 1 70 2 , 081 7.6 849.4 100 % PPC PZ 2 86 1 , 617.9 11.2 660.4 95 % PPC PZ 3 102 953.4 8.4 389.1 9 5 % PPC PZ 4 112 683.8 7.7 279.1 100% PPC PZ 5 120 614.8 8.0 250.9 100% PPC PZ6 154 188.3 4.2 76.9 100% PPC PPC: means the failure in the Polyester Polym er Concrete itself. Table 17 . T he total pull off test average of P 1 to P6 and PZ1 to PZ6. Steel PPC Panel Temperature Load (lbf) Force Rate (lbf/sec.) Tensile Stress (psi) Failure Mode P1&PZ 1 70 1 , 941.7 7.5 792.5 90 % PPC P2&PZ 2 86 1 , 667.1 11.4 680.5 89 % PPC P3&PZ 3 102 971.2 9.4 396.4 92.5 % PPC P4&PZ 4 112 727.9 8.7 297.1 100% PPC P5&PZ 5 120 630.9 7.9 257.5 100% PPC P6&PZ6 157 175.0 4.1 71.4 100% PPC PPC: means the failure in the Polyester Polymer Concrete itself.

PAGE 68

55 Compression Test The 2 00 Kips MTS machine was used for the PPC compression test because the load was expected to be very high, much t oo high for the 20 Kips MTS machine. As with the steel bond test, a procedure was set up. The procedure was setup for compression rather than tension. The force rate on the 2 0 0 Kips MTS was set at 100 pounds per second and a laser extensometer device was used to measure the cylinder compression strain, as shown in Figure 50 . The strain was calculated based on around 3.94 inches in length. After the 2 00 Kips MTS machine had been set up, the compression test was read y to be run ( Figure 50 ). When the test started running, displacement vs. time and force vs. time diagrams appeared on the MTS computer. The results from the MTS machine expressed force in terms of lbf units. This force was converted to strength expressed in psi units. The conversion was done by dividing the force from the MTS machine by the contact area between the MTS machine and the cylinder. To clarify, the following equations were used to determine the PPC comp ressive strength: Where: f' PPC : the PPC compre s sive strength that ha s been used in this paper ( psi ) P: the ultimate applied force from the MTS machine on a sample ( lbf ) A: the contact area between the MTS machine and the cylinder (in 2 )

PAGE 69

56 Figure 50 . T he laser extensometer that had been used for cylinder strain measurement .

PAGE 70

57 Compression Test Results All six of the cylinders were 328 days old. After having been prepared on March 2, 2107, they were tested on January 23, 2018. Figure 51 through Figure 56 show the stress strain diagram for the six cylinders, while Table 18 shows the results for each cylinder. The compression failure of the cylinders is shown in Figure 57 to Figure 62 . Table 18 . T he cylinder compression test results. Cylinder Load (lbf) Compression Stress (psi) Strain * Modulus of Elasticity (psi) Ultimate at 40% Ultimate at 40% C1 82,843 6,592.4 2,610.2 0.02043 0.00148 1,766,745 C2 78,555 6,251.2 --------------------C3 81,604 6,493.8 -----0.01985 ----------C4 79,532 6,329.0 2,531.6 0.02162 0.00213 1,188,544 C5 79,856 6,354.7 ---------------------C6 82,533 6,567.8 ---------------------Average 80,821 6,431.5 0.02063 1,477,644 The modulus at 40% of ultimate compressive strength (dividing the stress by the strain). * C2, C5 and C6 strain s could not be presented because there w as a problem with laser extensometer reads .

PAGE 71

58 Figure 51 . Stress strain diagram for C1 Figure 52 . Stress strain diagram for C 2 . Strain could not be measured see table 18 0.02043,6592.4 0 1000 2000 3000 4000 5000 6000 7000 0.000 0.005 0.010 0.015 0.020 0.025 Stress (psi) Strain (in/in) Cylinder C 1 6251.2 0 1000 2000 3000 4000 5000 6000 7000 0.000 0.005 0.010 0.015 0.020 0.025 Stress (psi) Strain (in/in) Cylinder C 2

PAGE 72

59 Figure 53 . Stress strain diagram for C 3 Figure 54 . Stress strain diagram for C 4 0.01985,6493.88 0 1000 2000 3000 4000 5000 6000 7000 0.000 0.005 0.010 0.015 0.020 0.025 Stress (psi) Strain (in/in) Cylinder C 3 0.02162,6329.0 0 1000 2000 3000 4000 5000 6000 7000 0.000 0.005 0.010 0.015 0.020 0.025 Stress (psi) Strain (in/in) Cylinder C 4

PAGE 73

60 Figure 55 . Stress strain diagram for C 5 . Strain could not be measured see table 18 Figure 56 . Stress strain diagram for C 6 . Strain could not be measured see tab le 18 6354.7 0 1000 2000 3000 4000 5000 6000 7000 0.000 0.005 0.010 0.015 0.020 0.025 Stress (psi) Strain (in/in) Cylinder C 5 6567.8 0 1000 2000 3000 4000 5000 6000 7000 0.000 0.005 0.010 0.015 0.020 0.025 Stress (psi) Strain (in/in) Cylinder C 6

PAGE 74

61 Figure 57 . The left picture showing C1 before the compression test starting and the picture on the right showing it after it failed. Figure 58 . The left picture showin g C 2 before the compression test starting and the picture on the right showing it after it failed.

PAGE 75

62 Figure 59 . The left picture showing C 3 before the compression test starting and the picture on the right showing it after it failed. Figure 60 . The left picture showing C 4 before the compression test starting and the picture on the right showing it after it failed.

PAGE 76

63 Figure 61 . The left pict ure showing C 5 before the compression test starting and the picture on the right showing it after it failed. Figure 62 . The left picture showing C 6 before the compression test starting and the picture on the right s howing it after it failed.

PAGE 77

64 CHAPTER V DISCUSSION Steel Bond Test Results Discussion This study investigated the bonds between sandblasted steel surfaces and two types of primer one without the other with zinc. These bonds were examined and evaluated by using SB and SBZ samples that were prepared using the methods of the fifth trial described in Chapter III. A comparison between the tensile strength of P1 and PZ1 and the bond strength of the SB and SBZ samples is shown in Figure 63 . Only P1 and P2 were included in this comparison because their temperature was approximately the same as that of the SB and SBZ samples -The bond strength of the steel to primer was clearly higher than the tensile strength steel bond samples. But, as is shown in Table 3 through Table 14 , failure sometimes occurred at the PPC primer interface. These failures would not be considered as steel bond failure because the steel strength was a minimum of 50 ksi. In c omparison, the steel bond strength in the samples using primer without zinc was 1.07 ksi and 1 . 3 9 ksi for the zinc primer. On the other hand, the PPC tensile strength was around 0.8 ksi, which was the minimal tensile strength. The zinc primer has higher bo nd strength than primer without zinc. The zinc primer average bond strength is 30% ( 319.2 psi ) higher than the primer without zinc average bond strength.

PAGE 78

65 Figure 63 . The average tensile strength e steel bond average strength. The bond failures shown in Figure 31 a, 31b, 31c and 31f , and Figure 32 a, 32b, and 32f could be considered as bonding failures, while the samp le in Figure 31 c appears to have had two force components on the bond surface. Perfect bond failures are shown in Figure 31 a and Figure 31 f. It is clear from looking at Figure 31 d and 31 e , and Figure 32 c, 32d and 32e that the failure occurred in a very small contact area between the steel and the prim er and then extended out through the middle of the primer layer. The debonding happened first and then the stress became significant around the debonding point -not uniform on the primer layer . This stress caused internal tension within the primer layer, which forced the failure to occur in the middle of the primer. The failures of these samples were considered as steel bond failures that demonstrate scenarios that may actually occur in the preparation PPC overlay for the steel decks of real world bridges. 1,069.6 1,388.7 735.6 849.4 500 700 900 1100 1300 1500 1700 1 2 3 4 5 6 7 8 9 10 11 Strength, psi Sample Number (Disc, SB or SBZ) SBs and SBZs Strength vs. P 1 and PZ 1 Cores Strength SB SBZ P1-70 PZ1-70 SBs Average SBZs Average P1-70 Average PZ1-70 Average

PAGE 79

66 Pull off Test Results Discussion Pull off test results were used to determine the tensile strength of the PPC at six and PZ1 to PZ6 with respect to the core temper curve on Figure 64 . (BF). There was one crack on each core that had BF. At temperatures hig failure (DF). The cores with ductile failure achieved their maximum tensile strength, which stayed constant until the failure happened. Typica lly, two or more cracks occurred on each core that had DF, as seen in Figure 65 . Based on the results of the pull off and PPC compressive strength tests., empirical equations were developed to calculate the tensile strength of the PPC at any environmental strength of each core was divided by the average compression stress of the PPC that was tested in this experimental work. T he experimental ratios of PPC tensile strength to compressive strength are graphed with environmental temperatures in Figure 66 . Statistics were used to develop two logistic regressions for the data shown in Figure 66 . Equation 1 uses the logistic regression statistics and PPC temperatures mentioned above to calculate the theoretical PPC tensile compressive ratio (k ALK ). Equation 2 uses that PPC tensile compressive ratio a The logistic regressions that had been used; First: the tensile strength cannot be zero. Second: the tensile strength can be zero

PAGE 80

67 comparison between R 70,t/c and k ALK is seen in Figure 67 . (Equation 1) (Equation 2) Where: R 70,t/c : The experimental PPC tensile strength ratio with respect to the PPC compressive f PPC,t : The PPC tensile strength (psi) f PPC,70 : k ALK : The theoretical PPC tensile strength ratio with respect to the PPC compressive strength a T: T

PAGE 81

68 Figure 64 . T he total average of all the cores r esults for the steel PPC panels bonded by primer without zin c and zinc primer. Figure 65 . More than one crack on the cores that had been heated above 0 100 200 300 400 500 600 700 800 900 1000 60 70 80 90 100 110 120 130 140 150 160 170 Tensile Strength, psi Total Pull Off Test Average (Primer Without Zinc & Zinc Primer) Total Results Average Primer Without Zinc Results Zinc Primer Results

PAGE 82

69 Figure 66 . T he total average of R 70,t/c for all cores results from the steel PPC panels b onded by primer without zinc and zinc primer. Figure 67 . R 70,t/c for all cores results and its average for the steel PPC panels bonded by primer without zinc and zinc primer. Also, the theoretical ratio k ALK 0% 2% 4% 6% 8% 10% 12% 14% 16% 60 70 80 90 100 110 120 130 140 150 160 170 R 70 ,t/c R 70 ,t/c vs. Temperature (Primer Without Zinc & Zinc Primer) Total Average Primer Without Zinc Zinc Primer 125.0 0% 2% 4% 6% 8% 10% 12% 14% 16% 60 70 80 90 100 110 120 130 140 150 160 170 R 70 ,t/c & k ALK R 70 ,t/c & k ALK vs. Temperature (Primer Without Zinc & Zinc Primer) Experimental Ratio Experimental Average Theoretical Ratio

PAGE 83

70 Compression Test Resu lts Discussion The PPC cylinders were tested in room temperature (70 F) 328 days after they had been cast in the cylinder molds. Table 18 shows the results of the cylinder compression failure at the room temperature, as shown in Figure 57 through Figure 62 . Equation 1 and Equation 2. Those equations can used to determine the PPC tensile strength at any PPC Conclusion PPC has been used as bridge deck overlay in the United States since the 1970s. The material has great comp ressive strength but, because of it its reaction to temperature variations, it may not be the best choice in all situations. This paper did not include detailed cost or value engineering for PPC. One advantage of PPC is its setting time, which averages abo ut two hours. Because of this short setting time, indirect costs due of the highway closure may decrease. This could make PPC a good choice for large, busy highways that carry high volumes of traffic. Protecting the environment is a significant concern whe n considering any construction material. Compared to bitumen, which is widely used as a bridge overlay, PPC would appear to be more environmentally friendly. Bitumen often comes lose and sticks to the exterior of vehicles, leading to the washing of those v ehicles, using water and chemicals. This does not occur when PPC is used as an overlay. The primer without zinc bond strength to sandblasted steel has an average strength of 1,069.6 psi, while the zinc primer bond strength is 1,388.7 psi.

PAGE 84

71 The tensile stren Typically, PPC tensile strength control led the pull off test results. The average compressive strength of the PPC cylinders 328 after they were cast was F uture Research The steel bond strength was not inspected under thermal conditions, which may i s subjected to horizontal shear. The tensile strength of the PPC was not examined by this study for any temperature below 70 F. The tensile strength of PPC under freezing conditions would be a good point for future study . The force rate is a significant factor in measuring tensile strength. A lower force rate corresponds to lower tensile strength, while a higher force rate corresponds to higher tensile strength. The relationship between the force rate and the tensile strength could not be developed in thi s paper because the force rate was random and by hand. PPC compressive strength was inspected on this paper only at 70 F. Thermal effects on the compressive strength of PPC would be a good topic for future study. The relationship between the PPC compressi ve strength and its tensile strength may be linear, but this cannot be known without further investigation.

PAGE 85

72 REFERENCES Albrecht, P., & Hall, T. T. (2003). Atmospheric Corrosion Resistance of Structural Steels. Journal of Materia ls in Civil Engineering , 15(1). Alethafa, M. (2016). Mechanical and Bonding Properties of Polyester Polymer Concrete Bridge Deck Overlay. Civil Engineering Department. Denver: University of Colorado Denver. Cunningham, W. C., Ramsey, R. A., & Autenrieth, R . E. (1989, March 28). United State Patent No. 4,816,503. Koch, G. H., Brongers, M. P., Thompson, N. G., Virmani, P., & Payer, J. (2002). Corrosion Costs and Preventive Strategies in the United States. U.S. Federal Highway Administration; NACE Internation al; CC Technologies Laboratories, Inc.; Case Western Reserve University.