Citation
Corrosion inhibitors for CFRP-strengthened steel members

Material Information

Title:
Corrosion inhibitors for CFRP-strengthened steel members
Creator:
Chai, Yufei ( author )
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
1 electronic file (125 pages) : ;

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:
Kim, Jimmy
Committee Members:
Rutz, Frederick
Nogueira, Carnot

Subjects

Subjects / Keywords:
Steel, Structural ( lcsh )
Carbon fiber-reinforced plastics ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Abstract:
This thesis focuses on zinc inhibitors used in steel specimens bonded with carbon fiber reinforced polymer (CFRP) sheets. Corrosion is one of the most heavily evaluated outcomes after completion of a construction period because a high level of corrosion on the formation of the structure may cause unforeseen consequences. The purpose of this thesis is to evaluate corrosion inhibitors' effect for CFRP-strengthened steel members. ( , )
Abstract:
The study includes corrosion simulation, steel strips testing, friction test and beam testing. Two Steel strips in a dimension of 100 mm long ×37 mm wide ×3 mm formed one specimen. Each specimen was bonded with CFRP sheets and the total number of 40 steel strips were tests. Load capacity of each specimen after corrosion exposure was tested. Beams without any protection, beams with zinc coins, and beams with zinc spray comprised the three categories of beams. After different time of corrosion exposure, load capacity of each beam was tested in bending tests. Corrosion rate of each category was calculated using the corrosion rate equation according to ASTM G1. From testing results summarized in tables and figures, zinc does some effect on inhibiting corrosion on steel members while between zinc coins and zinc spray, zinc spray worked better in resisting rust.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: Adobe Reader.
General Note:
Embargo ended 06/18/2018
Statement of Responsibility:
by Yufei Chai.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
on10225 ( NOTIS )
1022564422 ( OCLC )
on1022564422
Classification:
LD1193.E53 2017m C53 ( lcc )

Downloads

This item has the following downloads:


Full Text
CORROSION INHIBITORS FOR
CFRP-STRENGTHENED STEEL MEMBERS
by
YUFEI CHAI
B.S., Tianjin College, University of Science and Technology Beijing, 2012
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 2017
1


This thesis for the Master of Science degree by
Yufei Chai
has been approved for the
Civil Engineering Program
by
Jimmy Kim, Chair
Frederick Rutz
Carnot Nogueira
11


Chai, Yufei (M.S., Civil Engineering)
Corrosion Inhibitors for CFRP-Strengthened Steel Members Thesis directed by Professor Jimmy Kim.
ABSTRACT
This thesis focuses on zinc inhibitors used in steel specimens bonded with carbon fiber reinforced polymer (CFRP) sheets. Corrosion is one of the most heavily evaluated outcomes after completion of a construction period because a high level of corrosion on the formation of the structure may cause unforeseen consequences. The purpose of this thesis is to evaluate corrosion inhibitors’ effect for CFRP-strengthened steel members.
The study includes corrosion simulation, steel strips testing, friction test and beam testing. Two Steel strips in a dimension of 100 mm long X37 mm wide X3 mm formed one specimen. Each specimen was bonded with CFRP sheets and the total number of 40 steel strips were tests. Load capacity of each specimen after corrosion exposure was tested. Beams without any protection, beams with zinc coins, and beams with zinc spray comprised the three categories of beams. After different time of corrosion exposure, load capacity of each beam was tested in bending tests. Corrosion rate of each category was calculated using the corrosion rate equation according to ASTM Gl. From testing results summarized in tables and figures, zinc does some effect on inhibiting corrosion on steel members.
The form and content of this abstract are approved. I recommend its publication.
m
Approved: Jimmy Kim


ACKNOWLEDGMENTS
I would like to express my appreciation to all people for their participation in my thesis research process and helped me to get better testing results and motivated me to deal with technical obstacles when I was stuck in them. Thanks for my advisor Dr. Jimmy Kim, who discussed with me about my thesis topic at first and I became be interested in corrosion inhibitor after talking with him. And he motivated me to do more during the whole research. Experimental tests need trials and errors and every time when I met obstacle he was always the one gave me lots precious suggestions and supports. Thanks for Mr. Tom Thuis, as well as all stuff members who work in machine shop, they provided their supports and gave lots advices to me when I conducted my experimental tests and used testing machines. Thanks for Dr Nien-Yin Chang, he is always kind and I learned a lot from him too.
And great appreciation to Dr. Carnot Nogueira and Dr Frederick Rutz, who are the committee members in my defense team and they encouraged me to finish my research. I would like to express my appreciation to Ibrahim Bumadian, who assisted me a lot and guided me in experimental tests as well as other graduate students. Thanks for civil engineering department at university of Colorado Denver for proving me a chance to do a research that I was really interested in.
Besides, I need to say thank you to my family and my friends. I cannot be here without their supports.
IV


TABLE OF CONTENTS
I. Overview................................................................................1
1.1 Introduction.......................................................................1
1.2 Research Significance..............................................................2
1.3 Outline............................................................................2
II. Literature Review......................................................................4
2.1 Steel..............................................................................4
2.2 Fiber Reinforced Polymer (FRP).....................................................5
2.3 Carbon Fiber Reinforced Polymer (CFRP).............................................7
2.4 Corrosion and Types of Corrosion...................................................8
2.5 Electrochemical Reaction...........................................................9
2.6 Corrosion Inhibitors..............................................................10
2.7 The Rules in Selection of Inhibitors..............................................11
2.8 Zinc Corrosion Inhibitors.........................................................13
2.9 Materials.........................................................................13
III. Steel Strips Testing..................................................................18
3.1 Corrosion Simulation Tests for Steel Strips.......................................18
3.2 Tension Test......................................................................21
v


3.3 Friction Test...........................................................25
3.4 Coupon Test.............................................................27
IV. Beam Test...................................................................69
4.1 Corrosion Simulation....................................................69
4.2 Flexure Test............................................................71
V. Summary and Conclusions....................................................108
5.1 Summary and Conclusions................................................108
5.2 Recommendations for Future Research....................................108
REFERENCES.....................................................................110
vi


LIST OF TABLES
Table 3-1 Tensile capacity of strips without zinc coins......................................28
Table 3-2 Tensile capacity of strips with zinc protection....................................29
Table 3-3 Mass loss of strips without zinc coins.............................................30
Table 3-4 Mass loss of strips with zinc coins................................................31
Table 3-5 Mass loss of zinc coins............................................................32
Table 3-6 Corrosion rate of strips without zinc coins. (Based on ASTM G1 Standard Practice for
Preparing, Cleaning, and Evaluation Corrosion Test Specimens)...............................35
Table 3-7 Corrosion rate of strips with zinc under chloride corrosion. (Based on ASTM Standard
Practice for Preparing, Cleaning, and Evaluation Corrosion Test Specimens)..................37
Table 3-8 Inhibitor efficiency of zinc coins.................................38
Table 3-9 Friction test between CFRP sheet and steel strips surface without zinc............39
Table 3-10 Friction test between CFRP sheet and steel strips surface with zinc coins........40
Table 3-11 Tensile capacity of epoxy coupons (exposed to corrosion environment up to 144
hours).......................................................................................41
Table 3-12 Tensile capacity of CFRP coupons (exposed to corrosion environment up to 144
hours).......................................................................................42
Table 4-1 Load capacities of beams without any corrosion protection..........................73
Table 4-2 Load capacities of beams with zinc coins...........................................73
Table 4-3 Load capacities of beams with zinc spray...........................................73
Table 4-4 Mass loss of beams without corrosion protection....................................74
Table 4-5 Mass loss of beams corrosion with zinc coins.......................................74
Table 4-6 Mass loss of beams corrosion with zinc spray.......................................75
vii


Table 4-7 Corrosion rate of beams without zinc coins. (Based on ASTM G1 Standard Practice for Preparing, Cleaning, and Evaluation Corrosion Test Specimens)............................
89


LIST OF FIGURES
Figure 2-1 Chemical and Mechanical Properties of A992 Steel.............................15
Figure 2-2 Corrosion on the road.................................................................16
Figure 2-3 Corrosion on Brooklyn Bridge..................................................16
Figure 2-4 Metal activity sheet, (from Internet) ................................................17
Figure 3-1 Epoxy adhesive: (a) Masterbrace SAT 4500 Part A; (b) Masterbrace SAT 4500 Part
B........................................................................................43
Figure 3-2 Properties of epoxy...........................................................43
Figure 3-3 Battery charger used in tests.........................................................44
Figure 3-4 Schematic of test coupon......................................................44
Figure 3-5 Corrosion test for steel strips with zinc coins :(a) steel strips at 24 hr before test; (b)
48 hr; (c) 72 hr ; (d) steel strips at 96 hr before test; (e) steel strips at 120 hr before test.45
Figure 3-6 Corrosion in progress: (a) 24 hr; (b) 48 hr; (c) 72 hr; (d) 96 hr; (e) 120 hr; (f) 144 hr.
......................................................................................46
Figure 3-7 Corrosion process completed: (a) 24 hr; (b) 48 hr; (c) 72 hr; (d) 96 hr; (e) 120 hr; (f)
144 hr.................................................................................47
Figure 3-8 Zinc coins: (a) 0 hr; (b) 24 hr; (c) 48 hr; (d) 72 hr; (e) 96 hr............48
Figure 3-9 Mass loss of zinc coins.....................................................49
Figure 3-10 Tension test setup for strips with CFRP using MTS machine: (a) before loading (for w/o zinc at 0 hours with strain gauge); (b) after failure (for w zinc at 48 hours without strain
gauge)...................................................................................49
Figure 3-11 Strips with zinc and strips without zinc after corrosion: (a) 0 hr ; (b) 24 hr ;(c) 48 hr ; (d) 72 hr ; (e) 96 hr. (f) all strips.....................................................50
IX


Figure 3-12 Accelerate corrosion.................................................51
Figure 3-13 Load-CFRP strain of strip without zinc at 0 hrs......................51
Figure 3-14 Load-CFRP strain of strip without zinc at 24 hrs......................52
Figure 3-15 Load-CFRP strain of strip without zinc at 48 hrs......................52
Figure 3-16 Load-CFRP strain of strip without zinc at 72 hrs......................53
Figure 3-17 Load-CFRP strain of strip without zinc at 96 hrs......................53
Figure 3-18 Load-CFRP strain of strip with zinc at 0 hrs..........................54
Figure 3-19 Load-CFRP strain of strip with zinc at 24 hrs.........................54
Figure 3-20 Load-CFRP strain of strip with zinc at 48 hrs.........................55
Figure 3-21 Load-CFRP strain of strip with zinc at 72 hrs.........................55
Figure 3-22 Load-CFRP strain of strip with zinc at 96hr...........................56
Figure 3-23 Strips without zinc mass loss pattern: (a) strips mass loss in grams; (b) mass loss
percentage. (Mass loss percentage= (mass loss/original mass) * 100%).............57
Figure 3-24 Strips with zinc mass loss pattern: (a) strips mass loss in grams; (b) mass loss
percentage (Mass loss percentage= (mass loss/original mass) * 100%)..............58
Figure 3-25 Corrosion rate of steel strips without zinc coins....................59
Figure 3-26 Corrosion rate of steel strips with zinc coins.......................59
Figure 3-27 Comparison between the average ultimate load of strips without and with zinc
protection........................................................................60
Figure 3-28 Load-displacement of strips without zinc..............................60
Figure 3-29 Load-displacement of strips with zinc coins..........................61
Figure 3-30 Schematic of friction test............................................61
Figure 3-31 Friction test.........................................................62
x


Figure 3-32 Friction coefficient between CFRP sheet and steel without zinc protection after
corrosion.....................................................................................62
Figure 3-33 Friction coefficient between CFRP sheet and steel with zinc after corrosion......63
Figure 3-34 Coupons: (a) Epoxy; (b) CFRP.....................................................64
Figure 3-35 Epoxy coupons after conditioning: (a) Ohr. (b)24hr;(c) 48hr; (d)72hr; (e) 96hr;
(f)120hr......................................................................................65
Figure 3-36 Tension test: (a) epoxy coupon at failure; (b) laser extensometer................66
Figure 3-37 Tension test for CFRP coupons: (a) before tension, (b) after tension.............66
Figure 3-38 Epoxy coupons after tension test: (a) 0 hr; (b) 24 hr; (c) 48 hr; (d) 72 hr; (e) 96 hr; (f)
120 hr; (g) 144 hr...........................................................................67
Figure 3-39 CFRP coupons after tension test: (a) 0 hr; (b) 24 hr; (c) 48 hr; (d) 72 hr ; (e) 96 hr;
(f) 120 hr; (g) 144 hr........................................................................68
Figure 4-1 Beam bonded with CFRP and strain gauges...........................................89
Figure 4-2 Beam with zinc coins and depth of immersion.......................................77
Figure 4-3 Zinc Spray depth..................................................................77
Figure 4-4 Beam preparation...................................................................78
Figure 4-5 Setup for accelerated corrosion....................................................78
Figure 4-6 Beams after 48-hr corrosion exposure: (a)(b) Beams without protection; (c)(d) Beams
with zinc coins; (e)(f): Beams with zinc spray................................................79
Figure 4-7 Beams after 96-hr corrosion exposure: (a)(b) Beams without protection; (c)(d) Beams
with zinc coins; (e)(f): Beams with zinc spray................................................80
Figure 4-8 Beams after 144-hr corrosion exposure: (a)(b) Beams without protection; (c)(d)
Beams with zinc coins; (e)(f): Beams with zinc spray.........................................81
xi


Figure 4-9 Beams without any protection before corrosion simulation........................82
Figure 4-10 Beams with zinc coins before corrosion...........................................82
Figure 4-11 Beams with zinc spray protection before corrosion simulation.....................83
Figure 4-12 Beam deterioration...............................................................83
Figure 4-13 Test setup: (a) load cell and potentiometer; (b) PI gages........................84
Figure 4-14 Beam failure.....................................................................84
Figure 4-15 Control beam.....................................................................85
Figure 4-16 Beams without zinc after 48 hours’ corrosion exposure............................98
Figure 4-17 Beams with zinc coins after 48 hours’ corrosion exposure........................99
Figure 4-18 Beams with zinc spray after 48 hours’ corrosion exposure........................99
Figure 4-19 Beams without zinc after 96 hours’ corrosion exposure..........................87
Figure 4-20 Beams with zinc coins after 96 hours’ corrosion exposure........................87
Figure 4-21 Beams with zinc spray after 96 hours’ corrosion exposure........................88
Figure 4-22 Beams without protection after 144 hours’ corrosion exposure...................101
Figure 4-23 Beams with zinc coins after 144 hours’ corrosion exposure.......................102
Figure 4-24 Beams with zinc spray after 144 hours’ corrosion exposure.......................102
Figure 4-25 Load-Pi strain for beams without protection at 48 hrs: (a) BWOP1; (b) BWOP2. .. 90 Figure 4-26 Load-Pi strain for beams without protection at 96 hrs: (a) BWOP3; (b) BWOP4. .. 91 Figure 4-27 Load-Pi strain for beams without protection at 144 hrs: (a) BWOP5; (b) BWOP6. 92
Figure 4-28 Load-Pi strain for beams with zinc coins at 48 hrs: (a) BWZC1; (b) BWZC2.........93
Figure 4-29 Load-Pi strain for beams with zinc coins at 96 hrs: (a) BWZC3; (b) BWZC4.........94
Figure 4-30 Load-Pi strain for beams with zinc coins at 144 hrs: (a) BWZC5; (b) BWZC6......95
Figure 4-31 Load-Pi strain for beams with zinc spray at 48 hrs: (a) BWZS1; (b) BWZS2.......96
xii


Figure 4-32 Load-Pi strain for beams with zinc spray at 96 hrs: (a) BWZS3; (b) BWZS4..97
Figure 4-33 Load-Pi strain for beams with zinc spray at 144 hrs: (a) BWZS5; (b) BWZS6.98
Figure 4-34 Load-CFRP strain for beams without protection at 48 hrs...........99
Figure 4-35 Load-CFRP strain for beams without protection at 96 hrs...........100
Figure 4-36 Load-CFRP strain for beams without protection at 144 hrs..........101
Figure 4-37 Load-CFRP strain for beams with zinc coins at 48 hrs.....................102
Figure 4-38 Load-CFRP strain for beams with zinc coins at 96 hrs.....................103
Figure 4-39 Load-CFRP strain for beams with zinc coins at 144 hrs....................104
Figure 4-40 Load-CFRP strain for beams with zinc spray at 48 hrs.....................105
Figure 4-41 Load-CFRP for beams with zinc spray at 96 hrs............................106
Figure 4-42 Load-CFRP for beams with zinc spray at 144 hrs...........................107
xiii


CHAPTERI
OVERVIEW
1.1 Introduction
Corrosion is a common phenomenon. When structural materials that are prone to rust are used intensively as part of a structure or parts like beams or columns, corrosion could become a potential safety hazard because it deteriorates the load capacity of a structure as well as its service life, potentially causing unforeseen structural failures. Various kinds of construction materials exist, and engineers select proper materials from a list based on multiple factors such as the material’s properties. Among many popular construction materials, steel is deemed to be especially important. Steel can be built as a framed structure like steel bridges or it can be used in the reinforcement of concrete or works with other materials.
From a report published by the U.S. Department of Transportation, almost $276 billion was spent annually caused by corrosion between 1999 and 2001. The estimated cost includes the cost of protection, inspection, repair and removal of wasted materials (Cicek,2011,2014).
Preventing engineering materials from corrosion has become a much more urgent affair
due to several reasons. The approximate service life of an average residential and commercial
building is between 60 and 80 years. However, a higher demand for performance of construction
materials is often sought while a longer lifespan of construction materials is being pursued. Also,
the pollution caused by human activities around the world is speeding up the corrosion rate of
nearly all construction materials, resulting in a huge amount of waste. Additionally, incredibly
high population in metropolitan cities requires a higher standard for safety requirements and a
higher level of demand for construction of buildings. With new building codes are revised and
1


updated, lower service life due to severe corrosion on structural materials is not tolerable. Moreover, sustainable development has become a main trend in the modern society, and without a doubt, inefficient uses of construction materials makes sustainability much more difficult to achieve.
1.2 Research Significance
This thesis research addresses the effect of zinc inhibitors. Zinc coins and spraying zinc were used in steel strips and steel beams which were strengthened by bonding with CFRP sheets. Corrosion was stimulated on strips and beams at different time of corrosion exposure. Due to steel’s susceptibility to corrosion and its vital role in structural material market, the issue about how to prevent structural steel from rust must be taken into consideration.
Tension tests and bending tests were applied on un-rusted and rusted strips/beams respectively to illustrate the changing pattern of steel specimens. Comparisons among load capacities and the mass loss and corrosion rate of these specimens were made to show how zinc products (zinc coins and zinc spray) worked as corrosion inhibitors in chloride corrosion in a lab environment.
1.3 Outline
Five chapters are included in this report. In Chapter 1, introduction of corrosion is described and the reason why taking corrosion into research is necessary is followed. Chapter 2 is an introduction of corrosion engineering. Properties of steel type used in research and theories of steel corrosion, corrosion inhibitors and the mechanism of how corrosion inhibitors work will be discussed. And the reason why zinc was selected as corrosion inhibitors in tests will be discussed.
2


In Chapter 3, chloride corrosion exposure on steel strips which were bonded with CFRP sheets will be described. Materials and tools which were used to conduct electrochemical corrosion and in tensile tests are addressed. Procedures of conducting tensile tests and friction tests will be talked in detail. Apart from steel strips, epoxy coupons and CFRP coupons were prepared and being exposed in corrosion environment. Tests on all specimens were conducted. Calculation of corrosion rate of unprotected and protected steel strips before corrosion happens and after corrosion exposure is used to evaluate corrosion inhibitors’ effect (ASTM G1 2011). Load capacities of steel strips after different time of corrosion exposure were tested in tension tests.
Chapter 4 focuses on the procedures of conducting tests on steel beams. The preparation process of beam testing and the installation of corrosion stimulation process on beams and bending tests are addressed. Comparisons among bending tests’ results are talked. Chapter 5 presents a summary of all tests results and conclusions will be made.
3


CHAPTER II
LITERATURE REVIEW
2.1 Steel
Structural steel is one of the main building materials used in modern construction together with masonry, concrete, timber, etc. Structural steel is widely used in frame construction because of its outstanding advantages: first, when it is compared with timber, sometimes steel beam sections become lighter than timber beams when I beam is used. Although the density of structural steel is much higher than timber, which is 7850 kg/m3 when compared with 1190 kg/m3(hardwood). In construction with timber, thicker and larger beams should be designed and used, due to timber’s low density and stiffness, resulting in a higher self-weight of timber.
Secondly, steel pieces can be produced and premade so it saves time limit of a project. Unlike concrete-based buildings, concrete mixture process should be conducted in the construction field and it will need 7 to 28 days to finish curing before next steps. Thirdly, steel can save money due to its durability and sustainability. Steel components can be recycled and reused and it complies the sustainable development rules. And shorter time limit in for steel construction is economic in a long run. Structural steel is eco-friendly and with the popularity in recent years, green constructions based on steel materials sparked engineers’ interests gradually.
Among structural steel categories, ASTM A992 steel is one of the most commonly used carbon steel types in the steel beam production. It is used for the I and H shape beams and these two shapes of steel beams are widely used in bridge and building constructions. The chemical properties of A992 structural steel are listed in Figure 2-1. (ASTM A992 Beam)
4


2.2 Fiber Reinforced Polymer (FRP)
The history of using fiber reinforced polymer products in an industrial field has not even over than 100 years so it is a newly emerged material when compared to traditional construction materials.
When comes to FRP production history, FRP was firstly used in the aerospace field and then it became much more popular in military use around the 1960s. Then it attracted lots attention in chemistry world and automobile industry dominated the market after that. FRP entered the architectural industry in the 70s and civil engineers started to consider higher demands for materials stiffness, durability and anti-corrosion properties and FRP composites raised their interests. Then FRP composites started to play a key role in structural repair and rehabilitation. FRP products gained the popularity among new structural materials, however, the price of FRP laminates was relatively expensive in entire construction processes and the developing of its use in the structural world was tough at that moment. With increasing requirements of the dead load and live load in nowadays, FRP is widely used because of it provides strong strengthening effect especially on tension on structural members (Market reviews n.d, High-performance and specialty fibers 2016).
When it comes to fibers, there are various types of fibers used in FRP composites
production (Gowayed 2013). The components of varied materials always have different physical
or chemical properties or both. Composite them together give the opportunity to let dissimilar
materials work together to combine their advantages together and provide better service when load
applies. In the structure of fiber reinforced polymer composites, space structural high strength
material like fibers, rods are covered by another kind of isotropic material which usually has low
stiffness or weak in elasticity or toughness, which is called the matrix. The matrix keeps fibers/
5


rods in the place where they are supposed to be while in the manufacturing process and prevents high strength fibers from changing positions when carrying loads. The matrix which surrounds these fibers or rods makes them form into a whole body rather than in a form of cumulative layers of fibers or rods groups. (Gowayed 2013).
When comes to fibers, fibers in FRP composites are various. They could be made of any kinds of materials like ceramic or timber, etc. But among all kinds of FRP types, the most widely used fibers in forming FRP composites used in civil engineering are glass fibers and carbon fibers. High stiffness and strength that are provided from FRP composites enhance the loading capacity of structural components or they can be used as alternative materials in structural parts. FRP bars enjoy a lot of popularity in the reinforcement of concrete, they can be bond with concrete and increase the tensile capacities.
FRP products have been used in the civil engineering field since the last century because
of their many outstanding properties. First, FRP is light-weight and it has a high strength to weight
ratio so normally, engineers do not have a need to consider its self-weight in structural design
when compared its stiffness. This is unlike concrete, steel or other traditional structural materials
whose self-weight cannot be ignored. For example, load and resistance factor (LRFD) design
method used in the structural design, 1.2 times dead load (gravity loads from both permanent and
temporary elements) of a designed building should be taken into calculation. If the self-weight of
structural materials cannot be ignored and for most steel types, their density is around 7800kg/m3
so at most of the time, self-weight of steel must be taken into consideration in design. Even in ASD
(allowable strength design methods), dead loads of steel structure including self-weight cannot be
deleted from service load combinations. When self-weight of structural materials becomes a
problem, it makes the whole project much more challenging, especially in seismic zones. In
6


Seismic zones, soils get liquefied after earthquakes happen and the liquefaction decreases the bearing capacity of the foundation. The heavy self-weight of a structure would further decrease the bearing capacity.
Apart from increasing the performance of concrete, FRP bars make structural parts look better than steel bars in appearance after years of use. Steel bars are poor in resisting corrosion and corrosion-induced cracks penetration is subject to moisture content. Cracks propagated on concrete parts may bring more moisture into inner concrete particles and when water reaches steel bars, it makes the steel bars corrode faster. Worse outcomes may occur when the acid rain comes. Once severe corrosion happens to steel bars, they will lose their service ability and only concrete will remain for carrying loads. These will lead to the falling of concrete pieces and the exposure of steel bars in the air. Large amounts of investment will be needed for repair and rehabilitation. Contrastively, when FRP bars work as reinforcing bars, they are not sensitive to water content. FRP bars are corrosion resistant so they are much more durable than steel bars. Apart from this, FRP material has high stiffness and strength and they can be produced in different forms such as FRP laminates, FRP bars, etc. Civil engineers can choose one or more kinds of FRP products for building use.( The Society of Fiber Science and Techno, 2016).
2.3 Carbon Fiber Reinforced Polymer (CFRP)
CFRP can be treated as carbon reinforced plastic because it is similar to fiber glass. It is made of textile materials and it can be bonded with other classic construction materials like steel and concrete by using strong glue to increase their mechanical properties and thermal stability (Stratford 2008). CFRP can be produced in many forms to fulfill building requirements and unlike concrete, concrete usually needs 24 days curing time after pouring it into molds, CFRP sheets are
7


easy to be prepared and applied. It can be made in the form of sheets, engineers select proper types of CFRP products like rods and laminates according to required applications and other factors.
As a type of FRP products, CFRP has high strength, high stiffness and light self-weight and low density, and it has good resistance to corrosion (Song and Yu, 2015). It can be categorized into different groups based on modulus ranges. Among FRP products mentioned previously, CFRP has the highest stiffness and strength among FRP groups. Besides, it is easy to install on many structural elements like beams, columns, etc.
2.4 Corrosion and Types of Corrosion
From ASTM, corrosion usually happens on metals when deterioration happens between them and the surrounding environment (Cicek 2014, Corrosion Control-Cooling Systems). The deterioration can be caused by a chemical or electrochemical reaction.
Corrosion affects structural parts not only on their appearance but also on their mechanical properties like ductility and strength. There are many corrosion types, which include uniform attack corrosion, localized corrosion, atmospheric corrosion, galvanic corrosion, etc. these corrosion are categorized according to corrosion influenced area and corrosion mechanism.
Among corrosion types mentioned above, uniform attack corrosion is the most common one that happens almost everywhere. Uniform attack corrosion describes the deterioration of the whole body or the whole surfaces at the same time. The whole deterioration process will extend until the corrosion reaches a peak, which is also known where failure happens, and the load achieves at that moment when failure happens is called load capacity.
When comes to corrosion products normally formed on steel or iron surfaces, different kinds of corrosion products can be seen and the formation of them can be affected by temperature,
8


PH, oxygen contents, moisture content in the environment, etc. When there is no oxygen (vacuum), corrosion can hardly be formed on metallic materials. But it is difficult to ensure a 100 % vacuum environment in the construction field. When there is little oxygen, the speed of rust formation is slow. It turns out that iron cannot be fully oxidized and low oxygen content prevents the future formation of rust. When iron or steel is exposed to humid environment and oxygen is easily available, red-brown rusts which are mainly composed by Fe2C>3 can be found easily. Like in Figure 2-3, rust is everywhere on the bridge because the bridge is above the river so it is in an oxygen-abundant environment. Other chemicals in the air like SO2 in the air (a result of heavy pollution) speeds up the corrosion (Cicek 2014).
2.5 Electrochemical Reaction
Electrochemical reactions happen in the process when electrons transfer happens. Corrosion happens because differences of electrical potential exist on two bulks of metal (anode and cathode) in an electrically conductive environment. An electrolyte must exist for electron transmission between two sites.
When dissimilar metals appear at the same time, galvanic corrosion can be stimulated. One kind of metal corrodes faster than another and the forming of current in the system attributes to the deterioration of the properties of materials.
When corrosion happens, one site (metallic material) will lose electrons and the phenomenon that happens to this site is called metal oxidization. And the site where the oxidization happens called anode. Anode oxidation formation: M—>ne"+Mn+ (Corrosion Control-Cooling Systems). Electrons are released in the corrosion environment while the anodic metal becomes less stable after oxidation finishes. (Meyers n,d).
9


At the same time, another site of metal attracts electrons, which leads to a reduction reaction. And this procedure is called cathode reduction. (Mn+—>M). Metal atoms attract electrons which are produced because of oxidation reactions that happen on an anode. Normally, metal who plays a role as cathode becomes ‘stable’ after reduction than before. In the entire system, the number of electrons lost in the surface of oxidized metal equals to the number of electrons gained on the reduction site, which maintains the equilibrium of electrons during a chemical reaction.
Where there is a cathode reduction reaction, there may be different results depending on the respective material properties. A conductive wire appears when there is an anode oxidation and a cathode reaction. A salt-bridge may be formed in a galvanic cell as a result of half-cell reactions. Galvanic corrosion happens when a salt bridge formed. Iron atoms lose electrons when corrosion happens, which form an electron transport circulation in the corrosion environment.
Despite the basic requirements to form corrosion, oxygen content, water appearance, temperature, pH and other secondary factors affect corrosion results or even affect the formation of chemical corrosion products. In theory, the higher the moisture content, oxygen appearance, and temperature, the higher the corrosion rate will be in the system.
2.6 Corrosion Inhibitors
To prevent severe corrosion happens to engineer materials on structural main bodies, which may result in catastrophes by reducing structures’ service life or turning into sudden failures, different kinds of corrosion inhibitors are used in civil engineering field. Higher demands for restraining corrosion on structural materials are pursued because of heavy pollution.
There are two types of corrosion inhibitors applied in nowadays engineering industry. The first one is inorganic inhibitors and another is organic inhibitors. Among inorganic inhibitors, they
10


are divided into four types. Anodic corrosion inhibitors, cathodic inhibitors, mixed inhibitors which combine anodic inhibitors and cathodic inhibitors together. And the last one: volatile corrosion inhibitors (Myer 2005).
Anodic inhibitors reduced the rate of anodic reactions. When oxidation reaction happens on anodes, these inhibitors produce an oxide film on the metal surface which needs to be protected. And these are called anodic inhibitors. Passivation happens during the test after corrosion sediments settle and the oxidized layer which is made of deposits prevents metal which needs to be protected from being exposed to corrosion environment. Passivation makes oxygen and water become unavailable. Anodic inhibitors are usually called sacrificial anodes (Sacrificial Anodes n.d) because they corrode first during corrosion exposure (Cicek 2014).
For cathodic corrosion inhibitors, there are two ways to slow down the corrosion rate. In the first method, cathodic inhibitors decrease the cathodic reaction rate by using cathode poisons, and on the second way, selected deposits were generated on the cathodic areas. The appearance of deposits restrains the rate of cathodic reactions.
Anodic and cathodic corrosion inhibitors can be chosen into different tests and sometimes mixed inhibitors can reduce the rate of the anodic and cathodic reaction rate at the same time more efficiently.
2.7 The Rules in Selection of Inhibitors
When selecting corrosion inhibitors into tests, properties of the structural material which need to be protected became the first concern. Metal activity sheet became a dominant role in choosing inhibitors when different metals appear at the same time. The metal activity sheet shows the different activities undergo chemical reactions among types of metal. Comparisons on
11


reactiveness among different metals can be made by using the sheet. (Figure 2-4). The metal is less active than the ones listed above it on the metal activity sheet.
During an electrochemical reaction process, when anodic inhibitors are chosen, the part which is made of more reactive metal will work as the sacrificial anode and it will be exposed to corrosion. Sacrificial anodes are widely used in boating system when sea water works as an electrolyte (Wankhede n.d). And metal pieces which are made of zinc are chosen because zinc is more active than iron and it is economical to be used. The places where sacrificial anode attached decide the frequency of changing sacrificial anodes.
Because the sacrificial metal is much more anodic and active, it will lose more electrons than the original metal bulk which needs to be protected. When sacrificial metal starts to release electrodes, oxidization reaction will initiate. With the carrying on of oxidization reaction, corrosion is accumulated on the sacrificial anode. Ideally, corrosion happens only on sacrificial metal and the bulk metal should be well protected without being attacked by corrosion.
Apart from properties of construction materials, the problem how long the expected service life should be considered to avoid over-protecting or lacking protection. Methods and steps of inhibitors’ installation should be confirmed. For instance, when FRP sheets are used in concrete parts’ repair and rehabilitation, smoothness and cleanness of the applied surfaces should be ensured before applying FRP sheets on. Besides, other factors may affect the formation of corrosion products like local environment effects (like humidification, pH and frozen and thaw, etc.) and, safety requirements should be reconsidered as well.
12


2.8 Zinc Corrosion Inhibitors
From the metal activity series, when steel needs to be protected, almost all elements list beyond steel can work as corrosion inhibitors. Zinc can be selected as the sacrificial anode to protect steel products in the following thesis testing because the element zinc is more active than iron. Zinc is listed in a higher position than iron. Besides, zinc inhibitors are economical to use.
With the increasing of hydroxyl ions around the metal protected, the protected metal is passivated because of the products in electrochemical reactions prevent the metal being exposed to corrosion. Zinc coatings and zinc spraying are much more efficient in preventing corrosion accumulation than zinc coins or zinc blocks. There are two reasons for this. The first reason is that spraying zinc can function well as a thin barrier, which makes water and oxygen unavailable to metal pieces. Secondly, zinc spraying is fully covered on surfaces of metal pieces, which need to be protected. It will start to protect steel parts as soon as the electrochemical reaction starts. Protection becomes uniformly spread on steel members where zinc spray covers.
To reveal the use of zinc inhibitors applied in the structural field, four types of tests were conducted during the entire research. There were corrosion simulation tests on strips and beams, friction tests between steel strips and CFRP sheets after different time of corrosion exposure, tensile tests for steel strips, epoxy coupons and CFRP coupons. When comes to beam testing, bending tests were selected to crush beams and load capacity of each beam was tested.
2.9 Materials
All experimental tests conducted were based on the properties of ASTM A992 steel, type CF130 CFRP sheet and epoxy coupons. A992 steel is widely used in beam construction in bridges and buildings, and it is mainly used in producing I-beams and H-beams with a 345 MPa yield
13


strength and a 200 GPa (30,000,000 psi) elastic modulus. According to the product’s dash sheet, MBrace CFRP (CF-130) has an equivalent 0.165 mm thickness with a 240 GPa fiber modulus, which is relatively low when compared with other MBrace products like MBrace CF530, which has a 3800 MPa tensile strength and with the average thickness of 0.176 mm.
Fiber reinforced polymer is a newly emerged reinforcement material, and it can be produced in different forms and FRP sheets can be applied on the surface of cleaned structural members’ surface for strengthening. They can increase the load bearing capacity of beams, slabs and other structural members. Two parts of epoxy adhesives are used to glue CFRP sheets with steel surfaces. Epoxy resin adhesives fall into the non-Newtonian group (Optimizing Viscosity for Epoxy Adhesives n.d), which does not follow Newton’s Law of Viscosity with a low viscosity of 1600 cps at 75°F. Epoxy resin adhesives are best used in applications where the gap between two members is thin enough or the area that needs application is relatively small.
14


TYPICAL CHEMICAL PROPERTIES
Carbon. Max % 0.23
Manganese, % 0.50 to 1.60
Silicon. Max % 0.40
Vanadium, Max % 0.15
Columbium, Max % 0.05
Phosphorus, Max % 0.035
Sulfur, Max % 0.045
TYPICAL MECHANICAL PROPERTIES
Tensile Strength, ksi 65
Yield Point, ksi 65
Maximum Yield-to-Tensile Ratio 0.85
Figure 2-1 Chemical and Mechanical Properties of A992 Steel (ASTM A992 Beam n.d).
15


Figure 2-3 Corrosion on Brooklyn Bridge.
(https://commons.wikimedia.org/wiki/File:(Brooklyn_Bridge)_On_Corrosion,_Bridges_and_Sk
yscrapers.jpg, 2017)
16


potassium
sodium
calcium
magnesium
aluminium
carbon
zinc
iron
tin
lead
hydrogen
copper
silver
gold
platinum
most reactive
t
least reactive
K
Na
Ca
Mg
Al
C
Zn
Fe
Sn
Pb
H
Cu
Ag
Au
Pt
Figure 2-4 Metal activity sheet (The reactivity series of metals n.d)
17


CHAPTER III
STEEL STRIPS TESTING
3.1 Corrosion Simulation Tests for Steel Strips
There are four prerequisites that help form corrosion. During corrosion stimulation, both anode and cathode should exist at the same time. Secondly, electrolyte and the contact between cathode and anode (must be electrically conductive) should be fulfilled. The current in the system should return to the original source, and it is called a current return path. The changing of potentials on anodes and cathodes may result in current start to flow in another direction. These are called four ‘must haves’ to generate galvanic corrosion.
All materials used in tests should be prepared with caution because different processing procedures will affect the extent of corrosion sometimes even corrosion products. Steel and CFRP sheets are the main materials used in the present test. Steel strips with dimensions of 100 mm long X 37 mm wide X 3 mm thick were selected to use in steel strips testing. Two pieces of steel strips formed one test specimen. Before bonding CFRP sheets, the surface of each steel strip was cleaned by sand paper, a wire brush, and a grinder. Two reasons can explain why cleaning and polishing processes were necessary: firstly, rust existing on strips before wanted corrosion exposure may affect the final corrosion accumulation. Secondly, getting rid of rust before a corrosion simulation process begins makes a better bonding effect between steel substrates and CFRP sheets.
Type CF130 CFRP sheets were chosen, which have an intermediate fiber modulus 227 GPa and a tensile strength of 3,800 MPa (MBRACE® CF 130 n.d). Usually a single layer of CFRP sheet is in a nominal thickness of 0.165 mm according to products’ details. CFRP sheet pieces of
18


a dimension of 100 mm X 37 mm X 0.165 mm were used to strengthen steel strips, and they were
glued in the middle of two steel strips. Steel strips were bonded on both sides.
Zinc coins were selected as zinc inhibitors in tests for steel strips; they are all 31.75 mm +/- 1.59 mm in diameter and 2.54 mm in thickness. A blade saw machine was used here to cut zinc coins from zinc cast rods (all zinc cast rods came from rotometal.com). The unit weight of cast rod used in the test is 3.84 lbs. /ft., and they are 99.6% min zinc. The exact same type of zinc coins was applied in beam testing.
There are two parts of adhesives, which compose the epoxy resin used in the test. Epoxy is regarded as one kind of strong glue, and it is commonly used in mechanical bonding. Fiber reinforced polymer products (laminates, sheets, etc.) can be bonded to the surface of structural members, making the parts bond together tightly. The weight ratio of epoxy part A to part B is 3:1. Fine and quick mixing should ensure the quality of good gluing effect. Excessive glue should be removed, and an equivalent thickness of epoxy adhesive about 1mm was required. It took 7 days for the epoxy to be fully cured at room temperature (20°C). The properties of the epoxy resin are listed in Figure 3-2.
A gas torch was used to melt zinc coins. Welding was used to make zinc coins connect with steel strips without gaps, but it did not work well. That is why another method (using a gas torch) was selected. The melting ensured that the zinc coins’ surfaces firmly connected to the test specimens, which led to better electrical contact with steel and zinc coins when electrochemical corrosion happened. Tapes and plastic ties were used. They were used to attach zinc coins with steel strips’ surfaces. During corrosion exposure, it is possible that zinc coins corrode badly or the corrosion is uneven on entire surfaces of steel specimens. When this situation happens, tapes and
19


plastic bands will play key roles to prevent zinc coins from becoming isolated from steel strips’ surfaces.
Chloride corrosion: a 3.5% concentration NaCl solution was used during the whole corrosion simulation process. A normal concentration used in experimental tests is 3.5%, and it is used commonly as chemical media in studying corrosion on metals. Na+ and Cl- are both corpuscles, which are electrically conductive. The appearance of salt improved the conductivity of water, and water was the electrolyte in the corrosion test. The transfer of electrons and metal atoms were allowed because of the electrical conductivity of the electrolyte and the differences between electrical potential between anode and cathode. Electrical potential differences existing between dissimilar metals make sure galvanic corrosion happens.
Conductive wires were used to form anode and cathode connection. Anode (steel strips bonded with CFRP sheet) was connected to the positive side of the battery charger to make sure excessive corrosion happened to steel strips and the cathode linked to the negative side of the battery. The cathode can be any size of steel pieces and one 200 mm X 37 mm X 3 mm steel piece was selected. Same cleaning and polishing procedures were followed but on the cathode.
The connection stimulated producing rust on the steel strips bonded with CFRP surrounded by zinc coins. A battery charger provided input current into the reaction containers. It was used to accelerate the electrochemical corrosion process. For the reason, the formation of corrosion naturally is relatively slow and limited time allowed for a research conducts in the lab so that is why speeding up the corrosion is necessary. Voltage input on the strips boosts to generate the corrosion. The battery charger 1050-PE (charger 2/10/50 AMP) from Schumacher Electric was used in all corrosion stimulation tests at a current of 10 amp.
20


The volt-ampere meter is used to monitor the volt-ampere changing patterns when the corrosion propagating during the test. Ampere readings were measured every two hours at several random points on steel strips and the average values of the ampere readings were compared. Differences between the ampere readings from starting the test and to the at end of the test can be treated as an index to show the corrosion inhibitors’ effect.
Strain gauges were bonded to the CFRP sheet. The distance between the strain gauges was 15.875 mm from center to center and the first strain gauge located 19 mm from the center of the specimen. Strain gage readings developed when the specimen was loaded. With the tests carried, strains increased before the CFRP de-bonded and dropped after failure happened.
Normally, there are two methods to evaluate the effect of corrosion inhibitor. The first one is measuring the mass differences between metal pieces before tests and metal pieces after corrosion exposure. They were labeled under MBT (mass before corrosion exposure) and MAT (mass after corrosion exposure) in mass loss tables. Another method is using electrochemical measurements. Besides, microscopy can be used in a chemistry lab. In here, mass and area measurements were conducted to get mass loss curve and corrosion rate of each strip was calculated.
With the corrosion generating in containers during 96 hours’ exposure, the load capacity of each steel strip was decreasing while the mass loss of steel strips and the mass loss of zinc coins were increasing. The mass of zinc was measured too for each category.
3.2 Tension Test
Tensile tests were conducted on steel strips to reveal zinc coins’ ability working as corrosion inhibitors. From pictures taken during the whole corrosion exposure process, more and
21


more rust was formed with time counting. After 144 hours’ corrosion exposure, steel strips without zinc protection were seriously damaged. Some strips even cracked into two parts and they cannot be tested in tension tests. Steel strips with zinc coins looked better than steel strips without zinc coins even after 144 hours’ exposure. But only steel strips which were exposed in corrosion less than 96 hours were tested to make sure the same number of specimens were tested in both groups.
Two groups of steel strips were taken into research: Steel strips without corrosion inhibitors and steel strips with zinc coins’ protection in tensile tests and there are five categories steel strips after a different time of corrosion exposure (0 hours, 24 hours, 48 hours, 72 hours, and 96 hours respectively). The entire time length of corrosion exposure is 144 hours but heavy corrosion happened to steel strips which exposed in chloride corrosion more than 96 hours. Moreover, it became difficult to conduct tensile tests on these steel pieces, which were severely damaged because it was difficult to attach corroded steel pieces in tensile MTS machine. That is why only strips without corrosion, strips which were exposed in chloride corrosion environment less than 96 hours were taken into consideration for comparison here.
From the load capacities measured of steel strips without zinc coins, a decreasing curve of load capacity can be drawn. In the group steel strips without zinc coins, the average ultimate load capacity of steel strips was 23.4 kN when there was no corrosion happened, and it dropped to 8.3 kN after 24hours’ corrosion exposure. With the generating of rusts, the average load capacity dropped to 5.9 kN, 4.9 kN, 3.6 kN after 48hours, 72hours and 96hours’ corrosion exposure
22


respectively. Corrosion happened much more intense on the first 24-hour corrosion exposure from the decreasing curve.
The load capacity of steel strips with zinc coins’ protection was tested in tensile tests as well at different time categories and they are listed in Table 3-2. A similar decreasing trend line was witnessed during the entire corrosion tests, but by contrast, it showed some different results from steels strips without any protection: higher load capacities after corrosion tests when compared with strips without zinc coins. The failure load dropped from a same 23.4 kN to 8.7 kN after 24-hour chloride corrosion, and it decreased to an average value of 6.7 kN after 96 hours. Intense corrosion happened mostly at the first 24-hour corrosion exposure, but the final 6.7 kN that was near twice the load capacity of strips without zinc coins after 96 hours’ exposure. When compared with the group of steel strips without zinc protection, the average load capacity of steel strips with zinc coins’ protection decreased less. The decreasing pattern of load capacities shows zinc played a key role in preventing steel from a significant decrease of ultimate load capacity. It shows when zinc coins work as corrosion inhibitors in chloride corrosion test, it did some help in preventing rust accumulating. The corrosion rate of each steel strip and the efficiency of inhibitors were analyzed.
Mass loss of strips without zinc corrosion and strips without zinc protection were measured and the average mass loss of each category was calculated. Mass loss of each specimen was weighed after cleaning. The curve - which is the summary of mass loss of each specimen - is provided together with the percentage of average mass in Figure 3-23 and Figure 3-24. The mass of strips without zinc coins decreased more, which was a 20.72 g in average than strips with
23


protection (17.94 g in average) and this phenomenon complied with load capacity changing pattern. The more severe corrosion happened, the more mass loss was revealed.
The summary of the mass of each coin that measured after a different time of corrosion exposure was listed in Table 3-5. When making a comparison between these two tables, significant differences of mass loss can be drawn. The average mass loss of steel strips in two groups became similar in general. After 72 hours’ exposure, the difference between two groups became larger. For the group without zinc protection, the average mass loss was 6.76 g and the average mass loss of strips with zinc protection was 2.60g. A similar difference was found when compared with the average mass loss percentage. Zinc coins worked as sacrificial metals and the drop of mass showed their anti-corrosion effect. Zinc corroded first during chloride corrosion because it behaved much more reactive than steel as a result of its chemical properties. So, when corrosion was propagating, zinc coins became vulnerable and their mass decreased until the tests stopped. The cleaning process for zinc coins after corrosion before weighing became difficult because zinc coins became porous after being attacked by electrochemical corrosion. When the protection became less efficient, steel strips started to corrode. On each strip bonded with CFRP sheets applied in the test, there are four zinc coins (two on each side) attached to the steel surface. The mass loss of every coins was measured and then the average value was calculated.
When selecting zinc as the corrosion inhibitor in the tests above, it is obvious that zinc did some effects on protecting steel from rusting. The average corrosion rate of steel strips with zinc coins’ protection after 144-hour corrosion exposure was 11.594 cm/year and by contrast, the corrosion rate of steel strips without any corrosion protection was 14.342 cm/year. Significant differences of corrosion protection effect between unprotected steel and zinc-protected steel can be detected in a long term.
24


Standard deviation and coefficient of variation in math were used to evaluate the dispersion in a series of samples. Low standard deviation means almost all data approach to the mean value. After conducting tension tests on strips, load capacities of strips are data points used in dispersion analysis and the standard deviation was not low enough for the reason that data points are too limited to decide whether load capacities followed a standard deviation curve. The trend line was used to see the general changing pattern. Based on corrosion inhibitor efficiency in different categories listed in Table 3-8, it showed when zinc coins work as sacrificial anodes in corrosion exposure, it didn’t work perfectly, but they still decreased the corrosion rate somehow when they appeared around steel parts. Better connection between zinc coins and steel strips surfaces, more zinc coins or other types of zinc inhibitors like spraying zinc can be chosen to provide better corrosion protection.
3.3 Friction Test
The purpose of conducting friction tests on corroded steel strips was to reveal the anticorrosion effect of zinc coins not only from their appearances but also from physical properties. The more time of corrosion exposure on steel strips, the higher coefficient of friction was tested during friction tests. From the appearances of strips after being immerged in chloride corrosion environment for different time periods, strips became rougher as time went by. The coefficient of friction represents the frictional characteristics of tested materials. Before calculating the friction coefficient, the weight of each CFRP sheet was measured at 0 hours, 24hours, 48hours, 72hours and 96 hours respectively and the summary of all masses were listed in Table 3-9 and Table 3-10. Only four categories - (0 hour, 24 hour, 48 hour and 72 hour) of steel strips were tested in friction test because, steel strips corroded badly and only small pieces of them can be found after 72 hours’ corrosion exposure. Steel strips became shorter and limited surfaces area of steel strips will be
25


provided. As a consequence, the reading from friction test system was limited and became less accurate. A linear relationship between time and velocity can be drawn and the acceleration value can be calculated because it is the slope of the curve. According to Ffriction= P Nnominai, the coefficient of friction u was calculated. Average value of each u was summarized.
When performing the friction test, steel strips and CFRP coupons in each category were selected. A 50-mm diameter pulley system was the device to test coefficient of friction. A small weight was used to make sure CFRP sheet would slide on steel surfaces and sliding coefficient of friction could be tested. Steel strips were installed on a flat surface and a small weight of coin was attached to CFRP coupons’ surface using a tape and a thread. What needs to be paid attention to is that the weight should be selected properly: it should not be so large that it may pull the CFRP out of the steel strips’ surface while sliding happens, so there will be no reading on the software because of no contact between CFRP coupons and steel surface; On the other hand, if the weight is too small, it is likely that the weight cannot make CFRP coupons slide on the surface of steel strips, which may happen to steel strips that are heavily deteriorated.
The weight applied to the friction test were 0.00298 kg and 0.00594 kg. When steel strips were not seriously damaged, which the time of corrosion exposure was less than 24 hours, the weight of 0.00298 kg was heavy enough to slide the CFRP coupons on steel strips’ surfaces. The mass of each CFRP coupons was weighed before friction tests started. From the friction coefficient table, friction coefficient increased with the propagating of rust on steel strips, and when compared with steel strips without corrosion protection, the coefficient of friction was much lower on strips bonded with zinc coins. From the definition of friction coefficient, that means surfaces of steel
26


strips with zinc protection were much smoother than unprotected steel strips after corrosion
exposure.
3.4 Coupon Test
Epoxy coupons and CFRP coupons were made and then they were exposed to chloride corrosion environment together with steel strips to illustrate how corrosion influence their loading capacities. The dimension of each epoxy coupon was in a nominal length of 101.6 mm and a thickness of 5.08 mm. Epoxy resin was mixed and then poured in a mode, and they were ready to be exposed to chloride corrosion after 7 days curing time. The thickness of CFRP coupons were 0.165 mm and they were cut into 203.2 mm long and 12.5 mm wide segments. As same as steel strips testing groups, different time categories were made. Chloride corrosion was simulated on coupons for 0 hours, 24 hours, 48 hours, 72 hours, 96 hours, 120 hours and 144 hours respectively. The load capacity of each coupon was tested in tension test and the average value was calculated afterward. Load capacity of each epoxy coupon and CFRP coupon is listed in Table 3-11 and Table 3-12.
From Table 3-11, the average load capacity of epoxy coupons decreased from 2.10 kN to an average 2.04 kN after 144 hours’ corrosion exposure. The slight difference between these two average values means epoxy resin is not as vulnerable as steel member itself after chloride corrosion. For the reason that metal is much more electrical conductive, more electrochemical corrosion will happen on steel strips. For load capacities of CFRP coupons, no much differences between the average load capacities of coupons before corrosion and coupons after 144 hours was found. And this scenario approved that CFRP sheets were corrosion resistant.
27


Table 3-1 Tensile capacity of strips without zinc coins.
Without Ultimate Average Standard Coefficient of
Zinc Load Ultimate Deviation Variation
ID Load
(kN) (kN) (kN)
0-1 24.7
0-2 21.2 23.4 1.8 0.08
0-3 25.5
0-4 22.0
24-1 8.9
24-2 9.2 8.3 0.8 0.10
24-3 7.9
24-4 7.2
48-1 6.3
48-2 5.8 5.9 0.2 0.03
48-3 5.7
48-4 5.9
72-1 4.6
72-2 5.7 4.9 0.8 0.16
72-3 3.7
72-4 5.4
96-1 3.7
96-2 6.3 3.6 1.8 0.50
96-3 1.3
96-4 3.0
Coefficient of variation=standard deviation/average ultimate load
0-1: The first steel strip without zinc coins’ protection at 0 hour.
x-y: The yth steel strip without zinc coins’ protection after x hours’ corrosion exposure.
28


Table 3-2 Tensile capacity of strips with zinc protection.
With Ultimate Average Standard Coefficient of
Zinc Load Ultimate Deviation Variation
ID Load
(kN) (kN) (kN)
z-0-1 24.7
z-0-2 21.2 23.4 1.8 0.08
z-0-3 25.5
z-0-4 22.0
z-24-1 7.8
z-24-2 8.3 8.7 0.9 0.10
z-24-3 10.2
z-24-4 8.6
z-48-1 8.3
z-48-2 8.0 7.3 0.9 0.12
z-48-3 6.7
z-48-4 6.2
z-72-1 7.4
z-72-2 4.8 7.0 1.29 0.18
z-72-3 7.8
z-72-4 8.0
z-96-1 6.8
z-96-2 7.4 6.7 1.18 0.18
z-96-3 7.9
z-96-4 4.8
Coefficient of variation=standard deviation/average ultimate load
z-0-1: The first steel strip with zinc coins’ protection at 0 hour.
z-x-y: The yth steel strip with zinc coins’ protection after x hours’ corrosion exposure.
29


Table 3-3 Mass loss of strips without zinc coins.
ID MBT MAT ML Mass Avg ID MBT MAT ML Mass Avg
Loss Mass Loss Mass
Avg Loss Avg loss
(g) (g) (g) (g) (%) (g) (g) (g) (g) (%)
0 198.9 198.9 0.0 0.00 0.0
24-1 199.2 198.3 0.9 96-1 201.2 192.3 8.9
24-2 198.8 198.1 0.7 96-2 200.3 191.6 8.7
24-3 197.5 196.8 0.7 0.74 0.4 96-3 198.5 187.4 11.1 8.72 4.3
24-4 198.5 197.9 0.6 96-4 197.6 188.9 8.7
24-5 201.2 199.4 0.8 96-5 198.6 192.4 6.2
48-1 199.8 197.4 2.4 120-1 203.6 191.7 11.9
48-2 198.1 196.0 2.1 120-2 202.5 191.8 10.7
48-3 202.3 199.1 3.2 2.64 1.3 120-3 197.6 188.4 9.2 11.38 5.7
48-4 198.7 196.2 2.5 120-4 197.5 184.3 13.2
48-5 199.1 196.1 3.0 120-5 197.9 186.1 11.9
72-1 197.7 190.8 6.9 144-1 199.5 177.0 22.5
72-2 199.5 192.3 7.2 144-2 197.9 179.5 18.4
72-3 201.6 194.2 7.4 6.76 3.4 144-3 198.1 180.2 17.9 20.72 10.3
72-4 198.9 192.7 6.2 144-4 203.3 179.7 23.6
72-5 198.7 192.6 6.1 144-5 202.7 181.5 21.2
MBT: Mass before corrosion test.
MAT: Mass after corrosion test.
ML: Mass loss during the whole corrosion simulation process.
Aver Mass Loss (%) =average mass loss/average mass before tests.
30


Table 3-4 Mass loss of strips with zinc coins.
ID MBT MAT ML Mass Avg ID MBT MAT ML Mass Avg
Loss Mass Loss Mass
Avg Loss Avg loss
(g) (g) (g) (g) (%) (g) (g) (g) (g) (%)
0 198.9 198.9 0.0 0.00 0.0
24-1 199.1 198.7 0.4 96-1 201.3 196.7 4.6
24-2 200.1 199.6 0.5 96-2 197.6 192.0 5.6
24-3 199.4 199.1 0.3 0.42 0.2 96-3 198.0 192.2 5.8 5.24 2.6
24-4 189.5 189.2 0.3 96-4 198.0 193.2 4.8
24-5 192.6 192.0 0.6 96-5 201.3 195.9 5.4
48-1 200.1 199.5 0.6 120-1 200.9 185.0 15.9
48-2 197.9 197.2 0.7 120-2 199.3 192.1 6.8
48-3 200.2 198.6 1.6 0.96 0.5 120-3 198.2 188.8 9.4 9.98 5.0
48-4 198.1 196.5 1.6 120-4 197.8 191.4 6.4
48-5 189.9 189.6 0.3 120-5 198.1 176.7 11.4
72-1 197.3 194.6 2.7 144-1 198.2 186.6 11.6
72-2 201.3 197.6 3.7 144-2 199.0 178.0 21.0
72-3 199.1 196.8 2.3 2.60 1.3 144-3 197.0 176.9 20.1 17.94 9.0
72-4 197.9 195.9 2.0 144-4 200.4 173.8 26.6
72-5 198.8 196.5 2.3 144-5 200.1 189.7 10.4
MBT: Mass before corrosion test.
MAT: Mass after corrosion test.
ML: Mass loss during the whole corrosion simulation process.
Avg Mass Loss (%) =average mass loss/average mass before tests.
31


Table 3-5 Mass loss of zinc coins.
ID MBT (K) MAT (K) ML (K) Avg Mass Loss (K) ML (%) ID MBT (K) MAT (K) ML (K) Avg Mass Loss (K) ML (%)
0-1 15.7 15.7 0 0 0 96-1 15.2 11.2 4.0 5.96 39.5
0-2 14.6 14.6 0 96-2 14.7 10.1 4.6
0-3 16.0 16.0 0 96-3 16.8 13.9 2.7
0-4 15.3 15.3 0 96-4 16.5 5.3 10.2
0-5 15.5 15.5 0 96-5 15.7 10.0 5.7
0-6 15.0 15.0 0 96-6 14.8 12.3 2.5
0-7 14.8 14.8 0 96-7 14.5 10.2 4.3
0-8 15.7 15.7 0 96-8 13.7 9.5 6.2
0-9 14.8 14.8 0 96-9 16.1 5.0 10.1
0-10 16.3 16.3 0 96-10 15.6 7.0 8.6
0-11 13.8 13.8 0 96-11 15.3 8.9 6.4
0-12 15.5 15.5 0 96-12 15.2 7.4 7.8
0-13 16.4 16.4 0 96-13 14.4 5.7 8.7
0-14 17.0 17.0 0 96-14 14.9 10.9 4.0
0-15 16.0 16.0 0 96-15 15.3 10.0 5.3
0-16 15.5 15.5 0 96-16 13.9 6.6 7.3
0-17 13.5 13.5 0 96-17 14.4 10.0 4.4
0-18 12.5 12.5 0 96-18 14.8 10.2 4.6
0-19 13.0 13.0 0 96-19 14.8 7.5 7.3
0-20 10.5 10.5 0 96-20 15.2 10.7 4.5
24-1 15.7 13 2.7 1.67 11.0 120-1 15.1 11.7 3.4 6.86 45.4
24-2 14.6 13.2 1.4 120-2 14.9 6.2 8.7
24-3 16.0 14.5 1.5 120-3 16.8 8.0 8.8
24-4 15.3 13.9 1.4 120-4 16.5 10.2 6.3
24-5 15.5 13.8 1.5 120-5 15.2 7.1 8.1
24-6 15.0 13.2 1.8 120-6 14.8 8.9 5.7
24-7 14.8 13.5 1.3 120-7 15.3 9.7 5.6
24-8 15.7 14.5 1.2 120-8 14.7 6.0 8.7
24-9 14.8 12.9 1.9 120-9 15.5 12.9 2.6
24-10 16.3 14.6 1.7 120-10 16.0 8.8 7.2
24-11 13.8 12.0 1.8 120-11 15.2 7.6 7.6
24-12 15.5 14 1.5 120-12 14.6 6.1 8.5
24-13 16.4 14.6 1.8 120-13 13.5 5.5 8.0
32


24-14 17.0 15.1 1.9 120-14
24-15 16.0 14.6 1.4 120-15 Note:
24-16 15.5 14.2 1.3 120-16 (Jnly 13 zinc coins were found after test. 7 coins corroded harllv anrl pannrrf hp
24-17 13.5 12.0 1.5 120-17
24-18 12.5 10.9 1.6 120-18
24-19 13.0 11.2 1.8 120-19 found.
24-20 10.5 8.2 2.3 120-20
48-1 13.0 10.1 2.9 2.52 16.7 144-1 14.7 6.6 8.1 8.96 59.4
48-2 14.2 12.0 2.2 144-2 14.7 5.0 9.7
48-3 15.5 13.1 2.4 144-3 15.3 4.7 10.6
48-4 17.8 14.8 3.0 144-4 16.2 7.0 9.2
48-5 16.5 14.1 2.4 144-5 15.2 6.5 8.7
48-6 14.0 11.4 2.6 144-6 14.6 7.2 9.4
48-7 13.4 11.2 2.2 144-7 15.2 4.1 11.1
48-8 13.2 11.2 2.0 144-8 15.1 5.8 9.3
48-9 14.0 11.4 2.6 144-9 15.8 7.5 8.3
48-10 16.5 14.7 1.8 144-10 14.2 6.2 6.0
48-11 16.5 13.8 2.7 144-11 14.5 7.3 8.2
48-12 17.2 14.9 2.3
48-13 13.0 10.4 2.6
48-14 14.8 12.2 2.6
48-15 17.0 14.1 2.9
48-16 18.5 15.8 2.7
48-17 15.3 12.6 2.7
48-18 16.5 14.1 2.4
48-19 15.0 11.2 3.8
48-20 13.5 12.0 1.5
72-1 15.2 12.1 3.1 4.44 29.4 Note:
72-2 16.1 12.0 4.1
72-3 13.8 6.9 6.9 umy ii zinc coins were round
72-4 16.2 14.2 2.0 9 coins were missing.
72-5 17.3 14.9 2.7
72-6 15.4 14.2 3.2
72-7 15.5 13.9 1.6
72-8 17.2 10.4 6.8
72-9 14.8 6.0 8.8
72-10 15.0 6.6 8.4
72-11 14.7 10.5 4.2
72-12 14.5 10.5 4.0
72-13 15.2 5.0 10.2
72-14 16.8 14.3 2.5
33


72-15 13.2 11.7 1.5
72-16 15.1 9.8 5.3
72-17 14.4 10.2 4.2
72-18 14.7 12.3 2.4
72-19 15.4 10.9 4.5
72-20 13.9 11.5 2.4
MBT: M ass before corrosion test.
MAT: Mass after corrosion test.
ML: Mass loss due to corrosion.
Avg Mass Loss (%) =average mass loss/average mass before tests* 100%
34


Table 3-6 Corrosion rate of strips without zinc coins. (Based on ASTM G1 Standard Practice for
Preparing, Cleaning, and Evaluation Corrosion Test Specimens).
T (h) Corrosion Rate (cm/year) Avg Corrosion Rate (cm/year) Mass Loss (g) A (cm2)
0 0.000 0.000 0.0 162.220
24 2.619 0.9 159.782
24 2.036 0.7 159.846
24 2.037 2.153 0.7 159.772
24 1.746 0.6 159.738
24 2.327 0.8 159.884
48 3.563 2.4 156.604
48 3.097 2.1 157.662
48 4.736 3.921 3.2 157.068
48 3.705 2.5 156.882
48 4.506 3.0 154.784
72 7.039 6.9 151.924
72 7.364 7.2 151.544
72 7.589 6.960 7.4 151.130
72 6.347 6.2 151.396
72 6.460 6.1 146.346
96 6.915 8.9 149.602
96 6.825 8.7 148.166
96 8.663 6.812 11.1 148.944
96 6.819 8.7 148.306
96 4.838 6.2 148.964
120 8.924 11.9 124.006
120 8.087 10.7 123.044
120 8.181 8.823 9.2 104.572
120 9.963 13.2 123.206
120 8.959 11.9 123.524
144 15.905 22.5 109.626
144 12.818 18.4 111.238
144 11.641 14.342 17.9 119.166
144 17.513 23.6 104.432
144 13.836 21.2 118.744
T: Time of exposure in hours; Corrosion Rate= (K*W)/ (A*T*D)
K: A constant used in calculating corrosion rate. K = 87,600 in here. W: Mass loss in grams;
35


A: Surface area of strips in cm2; Measure the average length, width and thickness of each steel
strip. A=2*(length*width) +2*(length*thickness) +2*(width*thickness).
D: Density of the material (steel). For steel used in the tests, D = 7.85 g/cm3.
36


Table 3-7 Corrosion rate of strips with zinc under chloride corrosion. (Based on ASTM Standard
Practice for Preparing, Cleaning, and Evaluation Corrosion Test Specimens).
T (h) Corrosion Rate (cm/year) Avg Corrosion Rate (cm/year) W (g) A (cm2)
0 0.000 0.000 0.0 162.220
24 1.164 0.4 159.838
24 1.455 0.5 159.764
24 0.872 1.222 0.3 159.910
24 0.873 0.3 159.806
24 1.745 0.6 159.848
48 0.885 0.6 157.624
48 1.046 0.7 155.564
48 2.425 1.436 1.6 153.384
48 2.385 1.6 155.982
48 0.440 0.3 158.448
72 2.671 2.7 156.646
72 3.668 3.7 156.342
72 2.288 2.583 2.3 155.824
72 1.982 2.0 156.426
72 2.305 2.3 154.630
96 3.578 4.6 149.428
96 4.371 5.6 148.922
96 4.444 4.067 5.8 151.724
96 3.722 4.8 149.902
96 4.221 5.4 148.724
120 11.330 15.9 130.502
120 4.254 6.8 148.652
120 6.304 6.763 9.4 138.662
120 4.035 6.4 147.516
120 7.893 11.4 134.304
144 5.862 11.6 153.344
144 13.980 21.0 116.408
144 13.265 11.594 20.1 117.426
144 18.861 26.6 109.290
144 6.003 10.4 134.264
T: Time of exposure in hours;
Corrosion Rate= (K*W)/ (A*T*D)
K: A constant used in calculating corrosion rate. K= 87,600 in here. W: Mass loss in grams;
37


A: Surface area of strips in cm2; Measure the average length, width and thickness of each steel
strip. A=2*(length*width) +2*(length*thickness) +2*(width*thickness).
D: Density of the material (steel). For steel used in the tests, D = 7.85 g/cm3= 0.00785 g/mm3.
Table 3-8 Inhibitor efficiency of zinc coins.
ID Average Corrosion Rate ID Average Corrosion Rate Inhibitor Efficiency
(cm/year) (cm/year) (%)
24z 1.22 24n 2.15 43.24
48z 1.44 48n 3.92 63.38
72z 2.58 72n 6.96 62.89
96z 4.07 96n 6.81 40.30
120z 6.77 120n 8.82 23.35
144z 11.60 144n 14.32 19.16
24z, 48z, 72z, 96z, 120z, 144z: ID for steel strips with zinc coins’ protection in different category.
24n, 48n, 72n, 96n, 120n, 144n: ID for steel strips without zinc protection in each different category.
Inhibitor Efficiency (%) = 100*(CRunprotected — CRprotected)/ CRunprotected
38


Table 3-9 Friction test between CFRP sheet and steel strips surface without zinc.
T (hours) Weight (kg) Mcfrp (kg) Acceleration (m/s2) F (N) Friction Coefficient Avg Friction Coefficient
0 0.00298 0.0055 1.5145 0.004513 0.31
0 0.00298 0.0055 2.2825 0.006802 0.19
0 0.00298 0.0055 2.3225 0.006921 0.18 0.22
0 0.00298 0.0055 2.1638 0.006448 0.21
0 0.00298 0.0055 2.2548 0.006719 0.19
24 0.00298 0.0059 1.3608 0.004055 0.30
24 0.00298 0.0059 1.3702 0.004083 0.30
24 0.00298 0.0059 1.4885 0.004436 0.28 0.29
24 0.00298 0.0059 1.5646 0.004663 0.27
24 0.00298 0.0059 1.5709 0.004681 0.27
48 0.00594 0.0062 1.5542 0.009232 0.65
48 0.00594 0.0062 1.2552 0.007455 0.71
48 0.00594 0.0062 1.5499 0.009206 0.66 0.68
48 0.00594 0.0062 1.5846 0.009413 0.65
48 0.00594 0.0062 1.2661 0.007521 0.71
72 0.00594 0.0055 2.2734 0.013504 0.61
72 0.00594 0.0055 1.3129 0.007799 0.81
72 0.00594 0.0055 1.3399 0.007959 0.80 0.76
72 0.00594 0.0055 1.5274 0.009073 0.76
72 0.00594 0.0055 1.3485 0.008007 0.80
T: time.
Weight: weight of coins applied to cause the acceleration.
McfrP: weight of CFRP sheet.
Acceleration: Time-velocity linear curve can be drawn from the machine and acceleration is the slope of it.
F: normal force applied in the system. F = Weight * Acceleration
Friction coefficient = ((Weight *g) - (Weight + Mcfrp) * Acceleration))/ (Mcfrp * g)
g = 9.81 N/kg.
39


Table 3-10 Friction test between CFRP sheet and steel strips surface with zinc coins.
T (hours) Weight (kg) Mcfrp (kg) Acceleration (m/s2) F (N) Friction Coefficient Avg Friction Coefficient
0 0.00298 0.0055 2.3549 0.007018 0.18
0 0.00298 0.0055 2.5276 0.007532 0.15
0 0.00298 0.0055 2.1293 0.006345 0.21 0.21
0 0.00298 0.0055 1.9786 0.005896 0.24
0 0.00298 0.0055 1.8793 0.005600 0.25
24 0.00298 0.0059 1.6551 0.004932 0.26
24 0.00298 0.0059 1.9541 0.005823 0.21
24 0.00298 0.0059 1.7562 0.005233 0.24 0.24
24 0.00298 0.0059 1.7249 0.005140 0.25
24 0.00298 0.0059 1.8589 0.005540 0.23
48 0.00594 0.0062 1.8034 0.010712 0.60
48 0.00594 0.0062 1.6213 0.009631 0.64
48 0.00594 0.0062 1.7903 0.010634 0.61 0.61
48 0.00594 0.0062 1.7437 0.010358 0.62
48 0.00594 0.0062 1.8231 0.010829 0.60
72 0.00594 0.0055 1.7201 0.010217 0.72
72 0.00594 0.0055 1.6805 0.009982 0.73
72 0.00594 0.0055 1.7997 0.010690 0.71 0.72
72 0.00594 0.0055 1.8532 0.011008 0.69
72 0.00594 0.0055 1.6204 0.009625 0.74
T: time.
Weight: weight of hollow coins applied to cause the acceleration.
Mcfirp: weight of CFRP sheet.
Acceleration: Time-velocity linear curve can be drawn from the machine and acceleration is the slope of it.
F: normal force applied in the system. F = Weight * Acceleration
Friction coefficient = ((Weight *g) - (Weight + Mcfrp) * Acceleration))/ (Mcfrp * g)
G = 9.81 N /kg.
40


Table 3-11 Tensile capacity of epoxy coupons (exposed to corrosion environment up to 144 hours).
ID Failure Tensile Average Standard Coefficient
Load Strength Failure Deviation Of
Capacity Variation
(kN) (MPa) (kN/rMPal) (kN/rMPa])
0-1 2.0 24.3
0-2 2.9 34.8 2.10 0.47 0.22
0-3 1.8 21.4 [25.2] [5.75]
0-4 1.7 20.2
24-1 1.1 13.5
24-2 3.1 36.6 1.80 0.80 0.44
24-3 1.8 21.5 [21.5] [9.23]
24-4 1.2 14.5
48-1 2.9 34.6
48-2 3.5 42.0 3.15 0.54 0.17
48-3 2.4 28.4 [37.5] [6.48]
48-4 3.8 45.0
72-1 1.8 21.9
72-2 3.0 35.0 2.18 0.54 0.25
72-3 1.6 20.0 [26.1] [5.83]
72-4 2.2 27.5
96-1 2.0 23.6
96-2 1.7 20.0 1.9 0.12 0.06
96-3 2.0 23.3 [22.3] [1.43]
96-4 1.9 22.2
120-1 1.7 19.9
120-2 2.0 23.3 1.7 0.31 0.18
120-3 1.9 22.1 [19.9] [3.46]
120-4 1.2 14.3
144-1 1.3 15.1
144-2 2.3 27.7 2.04 0.48 0.24
144-3 2.6 31.0 [24.0] [5.96]
144-4 2.0 23.3
Cross-section area =12.7 mm (width) x 5.08mm (thickness) = 64.516 mm2
41


Table 3-12 Tensile capacity of CFRP coupons (exposed to corrosion environment up to 144 hours).
ID Failure Tensile Average Standard Coefficient
Load Strength Failure Deviation Of
Capacity Variation
(kN) (MPa) (kN/rMPal) (kN/rMPa])
0-1 8.8 4166.7
0-2 4.9 2320.1 5.9 1.7 0.29
0-3 5.2 2462.1 [2793.6] [797.2]
0-4 4.7 2225.4
24-1 4.9 2320.1
24-2 3.9 1846.6 4.8 0.7 0.14
24-3 5.8 2746.2 [2260.9] [327.0]
24-4 4.5 2130.7
48-1 4.7 2225.4
48-2 3.1 1467.8 4.8 1.4 0.28
48-3 4.5 2130.7 [2272.7] [644.0]
48-4 6.9 3267.0
72-1 5.5 2604.9
72-2 6.0 2840.9 4.5 1.3 0.30
72-3 3.9 1846.6 [2130.7] [636.1]
72-4 2.6 1231.1
96-1 5.4 2556.8
96-2 5.0 2367.4 5.0 0.2 0.05
96-3 4.8 2272.7 [2379.3] 107.8
96-4 4.9 2320.1
120-1 3.9 1846.6
120-2 2.9 1373.1 4.8 2.8 0.59
120-3 2.7 1278.4 [2260.9] [1336.4]
120-4 9.6 4545.5
144-1 3.7 1751.9
144-2 3.3 1562.5 5.2 1.7 0.34
144-3 6.2 2935.6 [2450.3] [825.1]
144-4 7.5 3551.1
Cross-section area= 12.7 mm (width) x 0.165 mm (thickness) = 2.0955 mm2
42


(a) (b)
Figure 3-1 Epoxy adhesive: (a) Masterbrace SAT 4500 Part A; (b) Masterbrace SAT 4500 Part B.
Technical Data
Product Chemistry MasterBrace® SAT 4500 Comp A MasterBrace® SAT 4500 Comp B Epoxy Resin Epoxy Hardener
Color Blue LX
Mixed Density 1,02 kg/litre
Viscosity 1500-2500 mPa.s
Compressive Strength TS EN 196 (7 days) > 60 N/mm2
Flexural Strength TS EN 196 (7 days) > 50 N/mm2
Bonding Strength to concrete (7 days) >3,0 N/mm2
Application Temperature +5°C - +30°C
Pot Life 30 minutes
Fully Cured at 20°C 7 days k*
Typical values are obtained from the test results of 4x4x16 mortar prism in 23°C and 50% relative humidity conditions. High temperatures shortens the curing and working time, lower temperatures extends the durations
Figure 3-2 Properties of epoxy.
43


Figure 3-3 Battery charger used in tests.
CL
44


(e)
Figure 3-5 Corrosion test for steel strips with zinc coins :(a) steel strips at 24 hr before test; (b) 48 hr; (c) 72 hr ; (d) steel strips at 96 hr before test; (e) steel strips at 120 hr before test.
45


(e) (f)
Figure 3-6 Corrosion in progress: (a) 24 hr; (b) 48 hr; (c) 72 hr; (d) 96 hr; (e) 120 hr; (f) 144 hr.
46


(e) (f)
Figure 3-7 Corrosion process completed: (a) 24 hr; (b) 48 hr; (c) 72 hr; (d) 96 hr; (e) 120 hr; (f) 144 hr.
47


(a) (b) _________________ (c)

(d) (e)
Figure 3-8 Zinc coins: (a) 0 hr; (b) 24 hr; (c) 48 hr; (d) 72 hr; (e) 96 hr.
48


12
10 -
3 8 -
(/)
(/)
° 6 -
(/)
(/)
0 4-
0
â–² â– 
A â– 
â–²
X <
A X
A i U
1 H â– 
A X
A â– 
I A â–  â– 


50
-------1-----------1-
100 150
Time (hours)
200
Figure 3-9 Mass loss of zinc coins.
(a) (b)
Figure 3-10 Tension test setup for strips with CFRP using MTS machine: (a) before loading (for w/o zinc at 0 hours with strain gauge); (b) after failure (for w zinc at 48 hours without strain gauge).
49


(e) (f)
Figure 3-11 Strips with zinc and strips without zinc after corrosion: (a) 0 hr ; (b) 24 hr ;(c) 48 hr ; (d) 72 hr ; (e) 96 hr. (f) all strips.
50


Load(kN)
Figure 3-12 Accelerate corrosion.
Figure 3-13 Load-CFRP strain of strip without zinc at 0 hrs.
51


Figure 3-14 Load-CFRP strain of strip without zinc at 24 hrs.
Figure 3-15 Load-CFRP strain of strip without zinc at 48 hrs.
52


Load (kN) Load/kN
Strain
Figure 3-16 Load-CFRP strain of strip without zinc at 72 hrs.
Figure 3-17 Load-CFRP strain of strip without zinc at 96 hrs.
53


Load (kN)
30 n
Figure 3-18 Load-CFRP strain of strip with zinc at 0 hrs.
Figure 3-19 Load-CFRP strain of strip with zinc at 24 hrs.
54


Load( KN) Load (kN)
7
Figure 3-20 Load-CFRP strain of strip with zinc at 48 hrs.
Figure 3-21 Load-CFRP strain of strip with zinc at 72 hrs.
55


Figure 3-22 Load-CFRP strain of strip with zinc at 96 hrs.
56


Time (hours)
(b)
Figure 3-23 Strips without zinc mass loss pattern: (a) strips mass loss in grams; (b) mass loss percentage. (Mass loss percentage= (mass loss/original mass) * 100%)
57


12 1
10 -
0 24 48 72 96 120 144
Time (hours)
(b)
Figure 3-24 Strips with zinc mass loss pattern: (a) strips mass loss in grams; (b) mass loss percentage (Mass loss percentage= (mass loss/original mass) * 100%).
58


Corrosion rate/(cm/year) Corrosion rate/(cm/year)
20 n
15 -10 -
5 -
0
0
t
â–²
2
â– 
â– 
â– 
â– 
l
~i---------------1---------------1
50 100 150
Time/(h)
Figure 3-25 Corrosion rate of steel strips without zinc coins.
59


Figure 3-27 Comparison between the average ultimate load of strips without and with zinc protection.
60



â–¡
Figure 3-30 Schematic of friction test.
61


Friction coefficient
Figure 3-31 Friction test.
Figure 3-32 Friction coefficient between CFRP sheet and steel without zinc protection after corrosion.
62


Friction coefficient
Figure 3-33 Friction coefficient between CFRP sheet and steel with zinc after corrosion.
63


CFRP Coupon Thickness=0 165
(b)
Figure 3-34 Coupons: (a) Epoxy; (b) CFRP.
64


(e) (f)
Figure 3-35 Epoxy coupons after conditioning: (a) Ohr. (b)24hr;(c) 48hr; (d)72hr; (e) 96hr; (f)120hr.
65


(a) (b)
Figure 3-36 Tension test: (a) epoxy coupon at failure; (b) laser extensometer.
66


(g)
Figure 3-38 Epoxy coupons after tension test: (a) 0 hr; (b) 24 hr; (c) 48 hr; (d) 72 hr; (e) 96 hr; (f) 120 hr; (g) 144 hr.
67


(g)
Figure 3-39 CFRP coupons after tension test: (a) 0 hr; (b) 24 hr; (c) 48 hr; (d) 72 hr ; (e) 96 hr; (f) 120 hr; (g) 144 hr.
68


CHAPTER IV
BEAM TEST
The chloride corrosion simulation accelerated the process of accumulating rust. After being exposed to corrosion, the purpose of bending tests on beams is to illustrate the de-bonding effect between CFRP sheets and after-corrosion steel beam’ surfaces. Beams in different categories were prepared, deteriorated and then tested under flexural tests.
Among structural members, beams take the responsibility of carrying loads together with columns and other structural elements. Beams carry transverse loading that means despite vertical loadings, end moments are carried by beams too. And among beams categorization, girders are the most vital beams in carrying loads, and bending is the main way to deflect a beam. As an important member in structural parts, the significance of preventing beams from corrosion is addressed.
A total of nineteen steel beams bonded with CFRP sheets were taken into research and they were processed in three groups: beams without any corrosion inhibitors’ protection, beams with zinc coins and beams with zinc anti-rusting spraying. There was one beam being tested as control beam and it was the beam without corrosion (no corrosion exposure). Other 18 beams are divided in three different corrosion exposure time: beams after 48 hours’ corrosion exposure, 96 hours’ corrosion exposure, and 144 hours’ exposure respectively.
4.1 Corrosion Simulation
A992 structural steel is normally used in I beam construction. Beams were prepared, applied corrosion and tested in bending tests. Wide flange I beam W 100 x 19 was cut by using a blade saw into 1000 mm-long beam segments. A slot was created on the one side on the flange by using the blade and the cutting depth was 30 mm from bottom flange. After being cut by the saw,
69


beams were cleaned and polished using a grinder to make sure no existing rusts. Then CFRP worked as a strengthening material to improve load capacities if beams under bending tests.
Same type of CFRP sheets were used for strengthening the bending capacity of steel beams. CFRP sheet was cut into 700 mm (L) X 100 mm (W) X 0.165 mm (T) segments. Epoxy adhesives were applied to ensure good bonding effect.
Zinc coins were cut from 31.75 mm +/- 1.59 mm in diameter and 2.54 mm in thickness from zinc cast rods. There were 18 zinc coins placed on bottom flange of each beam. Except beams bonded with zinc coins, another type of corrosion inhibitors being used in beam testing was spraying zinc. 3Mâ„¢ 16-501 Zinc Spray was selected in painting because of its high purity which was made of 97% pure zinc. The zinc spraying depth was 97 mm from the bottom flange (tension area). Firm and direct connection between zinc coins and steel surfaces is one of the necessary to prevent corrosion of the steel. Tapes and plastic ties were used to attach zinc coins with the surface of beams.
Similar chloride corrosion simulation procedures were continued as what had done in steel strips testing. Two beams were tested in each category. Every two beams which experienced same preparation were placed in a container and two timber-made blocks were placed underneath beams to make sure beams were not contact the bottom of container directly. To illustrate the corrosion inhibitors’ effect of zinc coins and zinc spray, beams were connected to the positive side of the charger and three pieces of steel strips in a dimension of 100 mm long X37 mm wide X3 mm connected to the negative side of the battery and were immerged in Nacl solution. The depth of immersion was 40 mm on each container. Two battery chargers provided input 2A ampere on each beam to start corrosion.
70


4.2 Flexure Test
One-point loading on middle span was applied gradually by a 20-kip MTS machine. The maximum displacement set for loading machine was 200 mm. Two-point supports were provided under the bottom flange of steel beams (which surface was bonded with CFRP sheet and would experience tension) in all bending tests and each support was located 50 mm away from the edge. Six strain gauges on each beam were organized and attached to the tension side of each beam on CFRP sheet for strain analysis while loading (see Figure 4-1). Load-strain curves were achieved after dealing with strain data. In theory, the maximum strain would happen near middle span that means strain gauges #3 or #4 (Figure 4-1) normally will tell the maximum strain because the steel fibers near middle span will experience the maximum elongation while tensile force applies. And either strain gauge #1 or #6 will experience the least strain among six strain gauges. From testing results (Figure 4-34 to Figure 4-42), the peak value of strain readings on #3 or #4 strain gauges located on the right of readings from other strain gauges. And this showed the small fiber particles attached near middle span experienced the largest elongation.
Pi gauges are Pi-shaped displacement transducers. The length of each gauge is 100mm and two gauges were used near the mid-span in two locations and they were used to test the strain of steel beam in a horizontal direction during the test. The pi gauge (labeled as #3) which was applied on top of steel beams experienced compression because it was installed in compression side while another one experienced tension (Pi gauge #5). When one beam experiences bending, the length of #3 gauge became smaller while the length of #5 became larger.
A load cell was installed beyond tested beams on the middle span before bending tests
conducted. And as a transducer, it recorded the load that beams experienced in seconds’ intervals
in flexural tests. A placement potentiometer was selected to be used in the test to record
71


displacement changing pattern of beams in the vertical direction. A load-displacement curve of each beam was listed in Figure 4-15 to Figure 4-24.
Based on testing results, beams without any corrosion protection dropped from 58 kN to an average value of 52.45 kN after 48-hour corrosion exposure and it dropped sharply to an average 46.45 kN and 40.05 kN after 96-hour and 144-hour corrosion exposure respectively.
High standard deviation shows the data is widely spread. Like beams without corrosion inhibitors, Load capacities of beams with zinc coins decreased with the deterioration of beams. After 144-hour corrosion exposure, the average load capacity of beams with zinc coins was 44.10 kN, which was higher than 40.05 kN which was the average reading of beams without protection. It showed zinc coins protected beams from rusting effectively to some extent. From Table 4-3, it showed a slightly decreasing pattern of load capacities on beams with spraying zinc protection. After 48-hour corrosion exposure with an average ultimate load 52.00 kN and there was only less than 2 kN difference between this group and the average ultimate capacity of beams which after 144-hour corrosion exposure.
Mass of beams before corrosion happened and after corrosion happened was weighed. They are summarized in Table 4-4 to Table 4-6 and they were listed under the label MBT. From tables, mass loss of beams with zinc coins and beams with zinc spray after corrosion exposure is much smaller than beams without any protection. It showed when zinc worked as inhibitors on steel members, it provides some helps to prevent heavy weight loss.
72


Table 4-1 Load capacities of beams without any corrosion protection.
Time of ID Ultimate Avg Standard Coefficient
Corrosion Load Ultimate Deviation of
Exposure Load Variation
(hour) (kN) (kN) (kN)
0 Control Beam 58.0 58.00 0.00 0.00
48 BWOP1 51.4 52.45 1.05 0.02
48 BWOP2 53.5
96 BWOP3 48.1 46.45 1.65 0.04
96 BWOP4 44.8
144 BWOP5 45.3 40.05 5.15 0.13
144 BWOP6 35.0
BWOP: Beams without corrosion protection (no corrosion inhibitors). Two beams were tested in each category.
Table 4-2 Load capacities of beams with zinc coins.
Time of ID Ultimate Avg Standard Coefficient
Corrosion Load Ultimate Deviation of
Exposure Load Variation
(hour) (kN) (kN) (kN)
0 Control Beam 58.0 58.00 0.00 0.00
48 BWZC1 50.0 50.70 0.70 0.01
48 BWZC2 51.4
96 BWZC3 48.3 49.15 0.85 0.02
96 BWZC4 50.0
144 BWZC5 41.6 44.10 2.50 0.06
144 BWZC6 46.6
BWZC: Beams with zinc coins.
Table 4-3 Load capacities of beams with zinc spray.
Time of ID Ultimate Avg Standard Coefficient
Corrosion Load Ultimate Deviation of
Exposure Load Variation
(hour) (kN) (kN) (kN)
0 Control Beam 58.0 58.00 0.00 0.00
48 BWZS1 50.5 52.00 1.50 0.03
48 BWZS2 53.5
96 BWZS3 55.4 54.55 0.85 0.02
96 BWZS4 53.7
144 BWZS5 48.6 50.65 2.05 0.04
144 BWZS6 52.7
BWZS: Beams with spraying zinc.
73


Table 4-4 Mass loss of beams without corrosion protection.
Time of ID MBT MAT ML Mass Avg
Corrosion Loss Mass Loss
Exposure (hour) (kg) (kg) (kg) Aveg (kg) (%)
48 BWOP1 19.14 19.14 0.00 0.01 0.05
48 BWOP2 19.00 18.98 0.02
96 BWOP3 19.12 19.10 0.02 0.02 0.11
96 BWOP4 18.90 18.88 0.02
144 BWOP5 19.06 19.04 0.02 0.03 0.16
144 BWOP6 19.18 19.14 0.04
BWOP: Beams without corrosion protection (no corrosion inhibitors). Two beams are tested in each category.
MBT: Mass before corrosion test.
MAT: Mass after corrosion test.
ML: Mass loss during the whole corrosion simulation process. Avg Mass Loss = (Mass loss average/average MBT) *100% Table 4-5 Mass loss of beams corrosion with zinc coins.
Time of ID MBT MAT ML Mass Avg
Corrosion Loss Mass Loss
Exposure (hour) (g) (g) (g) Avg (g) (%)
48 BWZC1 19.08 19.08 0.00 0.01 0.05
48 BWZC2 19.24 19.22 0.02
96 BWZC3 19.16 19.14 0.02 0.02 0.10
96 BWZC4 19.20 19.18 0.02
144 BWZC5 19.26 19.26 0.00 0.02 0.10
144 BWZC6 19.12 19.08 0.04
BWZC: Beams with corrosion inhibitors protection. (Zinc coins from cast rod). Two beams are tested in each category.
MBT: Mass before corrosion test.
MAT: Mass after corrosion test.
ML: Mass loss during the whole corrosion simulation process. Avg Mass Loss = (Mass loss average/average MBT) *100%
74


Table 4-6 Mass loss of beams corrosion with zinc spray.
Time of ID MBT MAT ML Mass Avg
Corrosion Loss Mass Loss
Exposure (hour) (kg) (kg) (kg) Avg (kg) (%)
48 BWZS1 19.12 19.10 0.02 0.01 0.00
48 BWZS2 19.36 19.36 0.00
96 BWZS3 19.14 19.12 0.02 0.02 0.10
96 BWZS4 19.12 19.10 0.02
144 BWZS5 19.14 19.12 0.02 0.02 0.10
144 BWZS6 19.10 19.08 0.02
BWZS: Beams with corrosion inhibitors protection. (Zinc spray from cast rod). Two beams are tested in each category.
MBT: Mass before corrosion test.
MAT: Mass after corrosion test.
ML: Mass loss during the whole corrosion simulation process. Avg Mass Loss = (Mass loss average/average MBT) *100%
75


Table 4-7 Corrosion rate of beams without zinc coins. (Based on ASTM G1 Standard Practice for Preparing, Cleaning, and Evaluation Corrosion Test Specimens).
Time Corrosion Rate Avg Corrosion Rate W Avg W A Avg A
(h) (mm/year) (mm/year) (g) (g) (mm2) (mm2)
48 0.000 1.913 0 10 607400 607800
48 3.825 20 608200
96 3.845 3.844 20 20 604600 604800
96 3.843 20 605000
144 2.577 3.864 20 30 601500 601650
144 5.151 40 601800
K: constant = 87600 in calcu ating corrosion rate.
D: Density of the material (steel) = 7.85 g/cm3 T: Time of exposure in hours;
Corrosion Rate = (K * W) / (A * T * D)
W: Mass loss in grams;
A: Surface area of beams in cm2; Measure the average length, width and thickness of each steel beam. A = 2 *(length * width) + 2 *(length * thickness) + 2 *(width * thickness).
76


40
J
Figure 4-2 Beam with zinc coins and depth of immersion.
Figure 4-3 Zinc Spray depth.
77


Figure 4-4 Beam preparation.
(c)
Figure 4-5 Setup for accelerated corrosion: (a) No corrosion; (b) after 24-hr corrosion exposure; (c)after 48-hr corrosion exposure.
78


(e) (f)
Figure 4-6 Beams after 48-hr corrosion exposure: (a) (b) Beams without protection; (c) (d) Beams with zinc coins; (e) (f): Beams with zinc spray.
79


80


(e) (f)
Figure 4-8 Beams after 144-hr corrosion exposure: (a) (b) Beams without protection; (c) (d) Beams with zinc coins; (e) (f): Beams with zinc spray.
81


Figure 4-9 Beams without any protection before corrosion simulation.
Figure 4-10 Beams with zinc coins before corrosion: (a) zinc coins, (b) overview.
82


Figure 4-11 Beams with zinc spray protection before corrosion simulation.
Figure 4-12 Beam deterioration.
83


Figure 4-13 Test setup: (a) load cell and potentiometer; (b) PI gages.
Figure 4-14 Beam failure.
84


Load (kN) Load(kN)
60
Displacement (mm)
Figure 4-15 Control beam.
Displacement (mm)
Figure 4-16 Beams without zinc after 48 hours’ corrosion exposure.
85


Load (kN) Load (kN)
60
Displacement (mm)
Figure 4-17 Beams with zinc coins after 48 hours’ corrosion exposure.
Displacement (mm)
Figure 4-18 Beams with zinc spray after 48 hours’ corrosion exposure.
86


Load(kN) Load(kN)
60
50 H
Displacement (mm)
Figure 4-19 Beams without zinc after 96 hours’ corrosion exposure.
Displacement (mm)
Figure 4-20 Beams with zinc coins after 96 hours’ corrosion exposure.
87


Full Text

PAGE 1

i CORROSION INHIBITORS FOR CFRP STRENGTHENED STEEL MEMBERS by YUFEI CHAI B.S., Tianjin College, University of Science and Technology Beijing, 2012 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfil lment of the requirements for the degree of Master of Science Civil Engineering 2017

PAGE 2

ii This thesis for the Master of Science degree by Yufei Chai has been approved for the Civil Engineering Program by Jimmy Kim , Chair Frederick Rutz Carnot Nogueira

PAGE 3

iii Chai, Yufei (M.S., Civil Engineering) Corrosion Inhibitors for CFRP Strengthened Steel Members Thesis directed by Pr ofessor Jimmy Kim. ABSTRACT T his thesis focus es on zinc inhibitors used in steel specimens bonded with carbon fiber reinforced polymer (CFRP) sheet s . Corrosion is one of the most heavily evaluated outcomes after completion of a construction period because a high level of corrosion on the formation of the structure may cause unforeseen consequences. The purpose of this thesis is to evaluate corrosion strengthened steel members. The study includes corrosion simulation, s teel strips testing, friction test and beam testing. Two Steel strips in a dimension of 100 mm long 37 mm wide 3 mm formed one specimen . E ach specimen was bonded with CFRP sheets and the total number of 40 steel strips were tests. Load capacity of each specimen aft er corrosion exposure was tested. Beams without any protection, beams with zinc coins, and beams with zinc spray comprised the three categories of beams. After different time of corrosion exposure, l oad capacity of each beam was tested in bending tests. Co rrosion rate of each category was calculated using the corrosion rate equation according to ASTM G1 . From testing results summarized in tables and figures, z inc does some effect on inhibiting corrosion on steel members . The form and content of this abstrac t are approved. I recommend its publication. Approved: Jimmy Kim

PAGE 4

iv ACKNOWLEDGMENTS I would like to expr ess my appreciation to all people for their participation in my thesis research process and helped me to get better testing results and motivated me to de al with technical obstacles when I was stuck in them. Thanks for my advisor D r. Jimmy Kim, who discussed with me about my thesis topic at first and I became be interested in corrosion inhibitor after talking with him . And he motivated me to do more during the whole research. Experimental tests need trials and errors and every time when I met obstacle he was always the one gave me lots precious suggestions and supports. Thanks for Mr. Tom Thuis, as well as all stuff members who work in machine shop, they pro vided their supports and gave lots advices to me when I conducted my experimental te sts and used testing machines . Thanks for Dr Nien Yin Chang, he is always ki nd and I learned a lot from him too. And great appreciation to D r . Carnot Nogueira and D r Frede rick Rutz , who are the committee members in my defense team and they encouraged me to finish my research . I would like to express my appreciation to Ibrah i m Bumadian , who assisted me a lot and guided me in experimental tests as well as other graduate stud ents . Thanks for civil engineering department at university of Colorado Denver for proving me a chance to do a research that I was really interested in. Besides, I need to say thank you to my family and my friends . I cannot be here without their supports .

PAGE 5

v TABLE OF CONTENTS I. Overview ................................ ................................ ................................ ................................ ..... 1 1.1 Introduction ................................ ................................ ................................ ...................... 1 1.2 Research Signi ficance ................................ ................................ ................................ ...... 2 1.3 Outline ................................ ................................ ................................ .............................. 2 II. Literature Review ................................ ................................ ................................ ....................... 4 2.1 Steel ................................ ................................ ................................ ................................ .. 4 2.2 Fiber Reinforced Polymer (FRP) ................................ ................................ ..................... 5 2.3 Carbon Fiber Reinforced Polymer (CFRP) ................................ ................................ ...... 7 2.4 Corrosion and Types of Corrosion ................................ ................................ ................... 8 2.5 Electrochemical Reaction ................................ ................................ ................................ . 9 2.6 Corrosion Inhibitors ................................ ................................ ................................ ....... 10 2.7 The Rules in Selection of Inhibitors ................................ ................................ ............... 11 2.8 Zinc Corrosion Inhibitors ................................ ................................ ............................... 13 2.9 Materials ................................ ................................ ................................ ......................... 13 III . Steel Strips Testing ................................ ................................ ................................ ................. 18 3.1 Corrosion Simulation Tests for Steel Strips ................................ ................................ ... 18 3.2 Tension Test ................................ ................................ ................................ ................... 21

PAGE 6

vi 3.3 Friction Test ................................ ................................ ................................ ................... 25 3.4 Coupon Test ................................ ................................ ................................ ................... 27 I V . Beam Test ................................ ................................ ................................ ............................... 69 4.1 Corrosion Simulation ................................ ................................ ................................ ..... 69 4.2 Flexure Test ................................ ................................ ................................ .................... 71 V . Summary and Conclusions ................................ ................................ ................................ ..... 108 5.1 Summary and Conclusions ................................ ................................ ........................... 108 5.2 Recommendations for Future Research ................................ ................................ ....... 108 REFERENCES ................................ ................................ ................................ ........................... 110

PAGE 7

vii LIST OF TABLES Table 3 1 Tensile ca pacity of strips without zinc coins. ................................ ............................... 28 Table 3 2 Tensile capacity of strips with zinc protection. ................................ ............................ 29 Table 3 3 Mass loss of strip s without zinc coins. ................................ ................................ ......... 30 Table 3 4 Mass loss of strips with zinc coins. ................................ ................................ .............. 31 Table 3 5 Mass loss of zinc coins. ................................ ................................ ................................ 32 Table 3 6 Corrosion rate of strips without zinc coins. (Based on ASTM G1 Standard Practice for Preparing, Cleaning, and Evaluation Corrosion Test Specimens). ................................ ............... 35 Table 3 7 Corrosion rate of strips with zinc under chloride corrosion. (Based on ASTM Standard Practice for Preparing, Cleaning, and Evaluation Corrosion Test Specimens). ........................... 37 Table 3 8 Inhibitor efficiency of zinc coins. ................................ ................................ ................. 38 Table 3 9 Friction test between CFRP sheet and steel strips surface without zinc. ..................... 39 T able 3 10 Friction test between CFRP sheet and steel strips surface with zinc coins. ............... 40 Table 3 11 Tensile capacity of epoxy coupons (exposed to corrosion environment up to 144 hours). ................................ ................................ ................................ ................................ ........... 41 Table 3 12 Tensile capacity of CFRP coupons (exposed to corrosion environment up to 144 hours). ................................ ................................ ................................ ................................ ........... 42 Table 4 1 Load capacities of beams w ithout any corrosion protection. ................................ ....... 73 Table 4 2 Load capacities of beams with zinc coins. ................................ ................................ ... 73 Table 4 3 Load capacities of beams with zinc spray. ................................ ................................ ... 73 Table 4 4 Mass loss of beams without corrosion protection. ................................ ....................... 74 Table 4 5 Mass loss of beams corrosion with zinc coi ns. ................................ ............................. 74 Table 4 6 Mass loss of beams corrosion with zinc spray. ................................ ............................ 75

PAGE 8

viii Table 4 7 Corrosion rate of beams without zinc coins. (Based on ASTM G1 Standard Practice for Preparing, Cleaning, and Evaluation Corrosion Test Specimens). ................................ ......... 89

PAGE 9

ix LIST OF FIGURES Figure 2 1 Chemical and Mechanical Properties of A992 Steel . ................................ .................. 15 Figure 2 2 Corrosion on the road. ................................ ................................ ................................ . 16 Figure 2 3 Corrosion on Brooklyn Bridge. ................................ ................................ ................... 16 Figure 2 4 Metal activity sheet. from Internet ................................ ................................ ...... 17 Figure 3 1 Epoxy adhesive: (a) Masterbrace SAT 4500 Part A; (b) Masterbrace SAT 4500 Part B. ................................ ................................ ................................ ................................ ................... 43 Figure 3 2 Properties of epoxy. ................................ ................................ ................................ .... 43 Figure 3 3 Battery charger used in tests. ................................ ................................ ...................... 44 Figure 3 4 Sc hematic of test coupon. ................................ ................................ ........................... 44 Figure 3 5 Corrosion test for steel strips with zinc coins :(a) steel strips at 24 hr before test; (b) 48 hr; (c) 72 hr ; (d) steel strips at 96 hr before test ; (e) steel strips at 120 hr before test. .......... 45 Figure 3 6 Corrosion in progress: (a) 24 hr; (b) 48 hr; (c) 72 hr; (d) 96 hr; (e) 120 hr; (f) 144 hr. ................................ ................................ ................................ ................................ ....................... 46 Figure 3 7 Corrosion process completed: (a) 24 hr; (b) 48 hr; (c) 72 hr; (d) 96 hr; (e) 120 hr; (f) 144 hr. ................................ ................................ ................................ ................................ ........... 47 Figure 3 8 Zinc coins: (a) 0 hr; (b) 24 hr; (c) 48 hr; (d) 72 hr; (e ) 96 hr. ................................ ..... 48 Figure 3 9 Mass loss of zinc coins. ................................ ................................ ............................... 49 Figure 3 10 Tension test setup for strips with CFRP using MTS machine: (a) befo re loading (for w/o zinc at 0 hours with strain gauge); (b) after failure (for w zinc at 48 hours without strain gauge). ................................ ................................ ................................ ................................ ........... 49 Figure 3 11 Strips with zinc and strips without zinc after corrosion: (a) 0 hr ; (b) 24 hr ;(c) 48 hr ; (d) 72 hr ; (e) 96 hr. (f) all strips. ................................ ................................ ................................ .. 50

PAGE 10

x Figure 3 12 Accelerate corrosion. ................................ ................................ ................................ . 51 Figure 3 13 Load CFRP strain of strip without zinc at 0 hrs. ................................ ...................... 51 Figure 3 14 Load CFRP strain of strip without zinc at 24 hrs. ................................ .................... 52 Figure 3 15 Load CFRP strain of strip without zinc at 48 hrs. ................................ .................... 52 Figure 3 16 Load CFRP strain of strip without zinc at 72 hrs. ................................ .................... 53 Figure 3 17 Load CFRP strain of strip without zinc at 96 hrs. ................................ .................... 53 Figure 3 18 Load CFRP strain of strip with zinc at 0 hrs. ................................ ............................ 54 Figure 3 19 Load CFRP strain of strip with zinc at 24 hrs. ................................ .......................... 54 Figure 3 20 Load CFRP strain of strip with zinc at 48 hrs. ................................ .......................... 55 Figure 3 21 Load CFRP strain of strip with zinc at 72 hrs. ................................ .......................... 55 Figure 3 22 Load CFRP strain of strip with zinc at 96hr. ................................ ............................ 56 Figure 3 23 Strips without zinc mass loss pattern: (a) strips mass loss in grams; (b) mass loss percentage. (Mass loss percentage= (mass loss/original mass) * 100%) ................................ ...... 57 Figur e 3 24 Strips with zinc mass loss pattern: (a) strips mass loss in grams; (b) mass loss percentage (Mass loss percentage= (mass loss/original mass) * 100%). ................................ ..... 58 Figure 3 25 Corrosion rate of ste el strips without zinc coins. ................................ ...................... 59 Figure 3 26 Corrosion rate of steel strips with zinc coins. ................................ ........................... 59 Figure 3 27 Comparison between the average ultimate load of strips without and with zinc protection. ................................ ................................ ................................ ................................ ..... 60 Figure 3 28 Load displacement of strips without zinc. ................................ ................................ 60 Fig ure 3 29 Load displacement of strips with zinc coins. ................................ ............................ 61 Figure 3 30 Schematic of friction test. ................................ ................................ ......................... 61 Figure 3 31 Friction test. ................................ ................................ ................................ ............... 62

PAGE 11

xi Figure 3 32 Friction coefficient between CFRP sheet and steel without zinc protection after corrosion. ................................ ................................ ................................ ................................ ...... 62 Figure 3 33 Friction coe fficient between CFRP sheet and steel with zinc after corrosion. ......... 63 Figure 3 34 Coupons: (a) Epoxy; (b) CFRP. ................................ ................................ ................ 64 Figure 3 35 Epoxy coupons after conditioning: (a) 0hr. (b)24hr;(c) 48hr; (d)72hr; (e) 96hr; (f)120hr. ................................ ................................ ................................ ................................ ........ 65 Figure 3 36 Tension test: (a) epoxy coupon at failure; (b) laser extensometer. ........................... 66 Figure 3 37 Tension test for CFRP coupons: (a) before tension. (b) after tension. ...................... 66 Figure 3 38 Epoxy coupons after tension test: (a) 0 hr; (b) 24 hr; ( c) 48 hr; (d) 72 hr; (e) 96 hr; (f) 120 hr; (g) 144 hr. ................................ ................................ ................................ ......................... 67 Figure 3 39 CFRP coupons after tension test: (a) 0 hr; (b) 24 hr; (c) 48 hr; (d) 72 hr ; (e) 96 hr; (f) 120 hr; (g) 144 hr. ................................ ................................ ................................ .................... 68 Figure 4 1 Beam bonded with CFRP and strain gauges. ................................ .............................. 89 Figure 4 2 Beam with zinc coins and depth of immersion. ................................ .......................... 77 Figure 4 3 Zinc Spray depth. ................................ ................................ ................................ ........ 77 Figure 4 4 Beam preparation. ................................ ................................ ................................ ....... 78 Figure 4 5 Setup for accelerated corrosion. ................................ ................................ .................. 78 Figure 4 6 Beams after 48 hr corrosion exposure: (a)(b) Beams without protection; (c)(d) Beams with zinc coins; (e)(f): Beams with zinc spray. ................................ ................................ ............ 79 Figure 4 7 Beams after 96 hr corrosion exposure: (a)(b) Beams without protection; (c)(d) Beams with zinc coins; (e)(f): Beams with zinc spray. ................................ ................................ ............ 80 Figure 4 8 Beams after 144 hr corrosion exposure: (a)(b) Beams without protection; (c)(d) Beams with zinc coins; (e)(f): Beams with zinc spray. ................................ ................................ 81

PAGE 12

xii Figure 4 9 Beams without any p rotection before corrosion simulation. ................................ ...... 82 Figure 4 10 Beams with zinc coins before corrosion. ................................ ................................ .. 82 Figure 4 11 Beams with zinc spray protection before corrosion simulation. ............................... 83 Figure 4 12 Beam deterioration. ................................ ................................ ................................ ... 83 Figure 4 13 Test setup: (a) load cell and potentiometer; (b) PI gages. ................................ ......... 84 Figure 4 14 Beam failure. ................................ ................................ ................................ ............. 84 Figure 4 15 Control beam. ................................ ................................ ................................ ............ 85 Figure 4 ................................ .......... 98 Figure 4 17 Beams with zinc c ................................ ...... 99 Figure 4 ................................ ..... 99 Figure 4 ................................ .......... 87 Figure 4 ................................ ...... 87 Figure 4 exposure. ................................ ..... 88 Figure 4 ............................ 101 Figure 4 . ................................ .. 102 Figure 4 ................................ . 102 Figure 4 25 Load Pi strain for beams without protection at 48 hrs: (a) BWOP1; (b) BWOP2. .. 90 Figure 4 26 Load Pi strain for beams without protection at 96 hrs: (a) BWOP3; (b) BWOP4. .. 91 Figure 4 27 Load Pi strain for beams without protection at 144 hrs: (a) BWOP5; (b) BWOP6. 92 Figure 4 28 Load Pi strain for beams with zinc coins at 48 hrs: (a) BWZC1; (b) BWZC2. ........ 93 Fig ure 4 29 Load Pi strain for beams with zinc coins at 96 hrs: (a) BWZC3; (b) BWZC4. ........ 94 Figure 4 30 Load Pi strain for beams with zinc coins at 144 hrs: (a) BWZC5; (b) BWZC6. ...... 95 Figure 4 31 Load Pi strain for beams with zinc spray at 48 hrs: (a) BWZS1; (b) BWZS2. ........ 96

PAGE 13

xiii Figure 4 32 Load Pi strain for beams with zinc spray at 96 hrs: (a) BWZS3; (b) BWZS4. ........ 97 Figure 4 33 Load Pi strain for beams with zinc spray at 144 hrs: (a) BWZS5; (b) BWZS6. ...... 98 Figure 4 34 Load CFRP strain for beams without protection at 48 hrs. ................................ ..... 99 Figure 4 35 Load CFRP strain for beams without protection at 96 hrs. ................................ .... 100 Figure 4 36 Load CFRP strain for beams without protection at 144 hrs. ................................ .. 101 Figure 4 37 Load CFRP strain for beams with zinc coins at 48 hrs. ................................ .......... 102 Figure 4 38 Load CFRP strain for beams with zinc coins at 96 hrs. ................................ .......... 103 Figure 4 39 Load CFRP strain for beams with zinc coins at 144 hrs. ................................ ........ 104 Figure 4 40 Load CFRP strain for beams with zinc spray at 48 hrs. ................................ ......... 105 Figure 4 41 Load CFRP for beams with zinc spray at 96 hrs. ................................ ................... 106 Figure 4 42 Load CFRP for beams with zinc spray at 144 hrs. ................................ ................. 107

PAGE 14

1 CHAPTER I OVERVIEW 1.1 Introduction Corrosion is a common phenomenon. When structural materials that are prone to rust a re used intensively as part of a structure or parts like beams or columns, corrosion could become a potential safety hazard because it deteriorates the load capacity of a structure as well as its service life, potentially causing unforeseen structural fail ures. Various kinds of construction materials exist , and engineers select proper materials from a list based on multiple factors such as the important. Steel can be built as a framed structure like steel bridges or it can be used in the reinforcement of concrete or works with other materials. From a report published by the U.S. Department of Transportation, almost $276 billion was spent annually caused by cor rosion between 1999 and 2001. T he estimated cost includes the cost of protection, inspection, repair a nd removal of wasted materials (Cicek, 2011,2014 ) . Preventing engineering materials from corrosion has become a much more urgent affair due to several reas ons. The approximate service life of an average residential and commercial building is between 60 and 80 years. However, a higher demand for performance of construction materials is often sought while a longer lifespan of construction materials is being pu rsued. Also, the pollution caused by human activities around the world is speeding up the corrosion rate of nearly all construction materials, resulting in a huge amount of waste. Additionally, incredibly high population in metropolitan cities requires a h igher standard for safety requirements and a higher level of demand for construction of buildings. With new building codes are revised and

PAGE 15

2 updated, lower service life due to severe corrosion on structur al materials is not tolerable. Moreover, sustainable d evelopment has become a main trend in the modern society, and without a doubt, inefficient uses of construction materials makes sustainability much more difficult to achieve. 1.2 Research S ignificance This thesis research addresses the effect of zinc in hibitors. Zinc coins and spraying zinc were used in steel strips and steel beams which were strengthened by bonding with CFRP sheets. Corrosion was stimulated on strips and beams at different time of corrosion exposure. Due to rosion and its vital role in structural material market, the issue about how to prevent structural steel from rust must be taken into consideration. Tension tests and bending tests were applied on un rusted and rusted strips/beams respectively to illustra te the changing pattern of steel specimens . Comparisons among load capacities and the mass loss and corrosion rate of these specimens were made to show how zinc products (zinc coins and zinc spray) worked as corrosion inhibitors in chloride corrosion in a lab environment. 1.3 Outline Five chap ters are included in this report . In Chapter 1, introduction of corrosion is described and the reason why taking corrosion into research is necessary is followed. Chap ter 2 is an introduction of corrosion engineering. Properties of steel type used in research and t heories of steel corrosion , corrosion inhibitors and the mechanism of how corrosion inhibitor s work will be discussed. And the reason why zinc was selected as corrosion inhibitors in tests will be discussed .

PAGE 16

3 I n C hapter 3, chloride corrosion exposure on steel strips which were bonded with CFRP sheets will be described. Materials and tools which were used to co nduct electrochemical corrosion and in tensile tests are addressed. Procedures of conducting tensile tes ts and friction tests will be talked in detail . Apart from steel strips, epoxy coupons and CFRP coupons were pre pared and being exposed in corrosion environment. Tests on all specimens were conducted. Calculat ion of corrosion rate of unprotected and protec ted steel strips before corrosion happens and after corrosion exposure (ASTM G1 2011) . Load capacities of steel strips after different time of corrosion exposure were tested in tension tests. Chapter 4 focu ses on the procedures of conducting tests on steel beams. The preparation process of beam testing and the installation of corrosion stimulation process on beams and bending tests are addressed . Comparisons among ben Chapter 5 presents a summary of all tests results and conclusions will be made.

PAGE 17

4 CHAPTER II LITERATURE REVIEW 2.1 Steel Structural steel is one of the main building materials used in modern construction together with masonry, concrete, timber, etc. Structural steel is widely used in frame construction because of its outstanding advantages: first, when it is compared with timber, sometimes steel beam sections become lighter than timber beams when I beam is used. Although the density of structural steel is much higher than timber, which is 7850 kg/m 3 when compared with 1190 kg/m 3 (hardwood). low density and stiffness, resulting in a higher self weight of timber. Seco ndly, steel pieces can be produced and pre made so it saves time limit of a project . Unlike concrete based buildings, concrete mixture process should be conducted in the construction field and it will need 7 to 28 days to finish curing before next steps . Th irdly, steel can save money due to its durability and sustainability . Steel components can be recycled and reused and it complies the sustainable development rules . And shorter time limit in for steel construction is economic in a long run. Structural stee l is eco friendly and with the popularity in recent years, Among structural steel categories , ASTM A992 steel is one of the most commonly used carbon steel type s in the st eel beam production. It is used for the I and H shape beams and these two shap e s of steel beams are widely used in bridge and building constructions. The chemical properties of A992 structural steel are listed in F igure 2 1 . ( ASTM A992 Beam )

PAGE 18

5 2.2 Fiber Rein forced Polymer (FRP) The history of using fiber reinforced polymer products in an industrial field has not even over than 100 years so it is a newly emerged material when compared to traditional construction materials . When comes to FRP production history, FRP was firstly used in the aerospace field and then it became much more popular in military use around the 1960s. Then it attracted lot s attention in chemistry world and a utomobile industry dominated the market after that. FRP entered the architectural i ndustry in the 70s and civil engineers started to consider higher demand s for materials stiffness, durability and anti corrosion properties and FRP composite s raised their interests. Then FRP composites started to play a key role in structural repair and r ehabilitation . FRP products gained the popularity among new structural mater ials, however , the price of FRP laminates was relative ly expensive i n entire construction processes and the developing of its use in the structural world was tough at that moment . With increasing requirements of the dead load and live load in nowadays, FRP is widely used because of it provides strong strengthening effect especially on tension on structural members (Market reviews n.d , High performance and specialty fibers 2016). Whe n it comes to fibers, there are various types of fibers used in FRP composites production (Gowayed 2013) . The components of varied materials always have different physical or chemical properties or both. Composite the m together give the opportunity to let dissimilar materials work together to combine their advantages together and provid e better service when load applies . In the structure of fiber reinforced polymer composites, space structural high strength material like fibers, rods are covered by another kind of isotropic material which usually ha s low stiffness or weak in elasticity or toughness , which is called the matrix. The matrix keep s fibers /

PAGE 19

6 rods in the place where they are supposed to be while in the manufacturing process and prevent s high streng th fibers from changing positions when carrying load s . The matrix which surround s these fibers or rods make s them form into a whole body rather than in a form of cumulative layers of fibers or rods groups. (Gowayed 2013 ). When comes to fibers, f ibers in F RP composites are various. T hey could be made of any kinds of materials like ceramic or timber, etc. But among all kinds of FRP types, the most widely used fibers in forming FRP composites used in civil engineering are glass fibers and carbon fibers. High stiffness and strength that are provided from FRP composites enhance the loading capacity of structural components or they can be used as alternative materials in structural parts . FRP bars enjoy a lot of popularity in the reinforcement of concrete, they c an be bond with concrete and increase the tensile capacities. FRP products have been used in the civil engineering field since the last century because of their many outstanding properties. First, FRP is light weight and it has a high stren gth to weight ra tio so normally, engineers do not have a need to consider its self weight in structural design when compared its stiffness. This is u nlike concrete, steel or other traditional structural materials whose self weight cannot be ignored. For example, load and resistance factor (LRFD) design method used in the structural design , 1.2 times dead load (gravity loads from both permanent and temporary elements) of a designed building s hould be taken into calculation. I f the self weight of structural materials cannot be ignored and for most steel types, their density is around 7800kg/m3 so at most of the time, self weight of steel must be taken into consideration in design . Even in ASD (allowable strength design methods), dead loads of steel structure including self we ight can not be deleted from service load combinations. When self weight of structural materials becomes a problem, it makes the whol e project much more challenging, especially in seismic zones. In

PAGE 20

7 Seismic zones , soils get liquefied after earthquakes happen and the liquefaction decreases the bearing capacity of the foundation. The heavy self weight of a structure would further decrease the bearing capacity. Apart from increasing the performance of concrete, FRP bars make structural parts look better than ste el bar s in appearance after years of use . Steel bars are poor in re sisting corrosion and corrosion induced cracks penetration is subject to moisture content. Cracks propagated on concrete parts may bring more moisture into inner concrete particles and when water reaches steel bars, it ma kes the steel bars corrode faster. Worse outcomes may occur when the acid rain comes. Once severe corrosion happens to steel bars, they will lose their service abil ity and only concrete will remain for carry ing loads . These will lead to the falling of concrete pieces and the exposure of steel bars in the air. Large amount s of investment will be needed for repair and rehabilitation . Contrastively, w hen FRP bars work as reinforcing bars, they are not sensitive to water content . FRP bars are corrosion resistant so they are much more durable than steel bars . Apart from this, FRP material has high stiffness and strength and t hey can be produced in different forms such as FRP laminates, FRP bars, etc . Civil engineers can choose one or more kinds of FRP products for building use. ( The Society of Fi ber Science and Techno, 2016). 2.3 Carbon Fiber Reinforced Polymer ( CFRP ) CFRP can be treated as carbon reinforced plastic because it is similar to fiber glass. It is made of textile materi als and it can be bonded with other classic construction materials like steel and concrete by using strong glue to increase their mechanical properties and thermal stabilit y ( Stratford 2008 ). CFRP can be produced in many forms to fulfill building requireme nts and unlike concrete, concrete usually needs 24 days curing time after pouring it in to molds , CFRP sheets are

PAGE 21

8 eas y to be prepared and applied . It can be made in the form of sheets , engineers select proper types of CFRP products like rods and laminates a ccording to required applications and other factors . As a type of FRP products, CFRP has high strength, high stiffness and light self weight and low density , and it has good resistance to corrosion ( Song and Yu, 2015) . It can be categorized into different groups based on modulus ranges . Among FRP products mentioned previous ly , CFRP has the highest stiffness and strength among FRP groups . Besides, it is easy to install on many structural elements like beams, columns, etc. 2.4 Corrosion and Types of Corrosio n From ASTM, corrosion usually happens on metals when deterioration happens between them and the surrounding environment (Cicek 2014 , Corrosion Control Cooling Systems ). The deterioration can be caused by a chemical or electrochemical reaction. Corrosio n a ffects structural parts not only on their appearance but also on their mechanic al properties like ductility and strength . There are many corrosion types , which include uniform attack corrosion, localized corrosion, atmospheric corrosion , galvanic corrosion, etc. these corrosion are categorized according to corrosion influenced area and corrosion mechanism. Among corrosion types mentioned above, uniform attack corrosion is the most common one that happens almost everywhere . Uniform attack corrosio n describes the deterioration of the whole body or the whole surfaces at the same time. The whole d eterioration process will extend until the corrosion reaches a peak, which is also known where failure happens, and the load achieves at that moment when fai lure happens is called load capacity . When comes to corrosion products normally formed on steel or iron surfaces, different kinds of corrosion products can be seen and the formation of them can be affected by temperature,

PAGE 22

9 PH, oxygen contents, moisture co ntent in the environment, etc . When there is no oxygen (vacu um), corrosion can hardly be formed on metallic materials. But it is difficult to ensure a 100 % vacuum environment in the construction field. When there is little oxygen, the speed of rust format ion is slow. It turns out that i ron cannot be fully oxidized and low oxygen content prevent s the future formation of rust . When iron or steel is exposed to humid environment and oxygen is easily available, red brown rusts which are mainly composed by Fe 2 O 3 ca n be found easily. Like in F igure 2 3, rust is everywhere on the bridge because the bridge is above the river so it is in an oxygen abundant environment . Other chemicals in the air like SO 2 in the air ( a result of heavy pollution) speeds up the corrosio n ( Cicek 2014). 2.5 Electrochemical R eaction Electrochemical reaction s happen in the process when electrons transfer happens. Corrosion happens because differences of electrical potential exist on two bulks of metal (anod e and cathode) in an electrically conductive environment. An ele ctrolyte must exist for electron transmission between two sites. When dissimilar metal s appear at the same time , galvanic corrosion can be st imulated. O ne kind of metal corrode s faster than another and the forming of current in the system attributes to the deterioration of the properties of materials . When corrosion happens, one site (metal lic material ) will lose electrons and the phenomenon that happen s to this site is called metal oxidization. And the site where the oxidiza tion +M n+ ( Corrosion Control Cooling Systems ) . Electrons are released in the corrosion environment while the anodic metal becomes less stable after oxidation finishes . ( Meyers n,d ).

PAGE 23

10 At the same time, an other site of metal attracts electrons, which leads to a reduction reaction. And this procedure is called cathode reduction. (M n+ . Metal atoms attract electrons which are produced because of oxid ation reactions that happen on an anode. Normally, metal w ho than before. In the entire system , the number of electrons lost in the surface of oxidized metal equals to the number of electrons gain ed on the reduction site, which maintains the equilibrium of electrons during a chemical reaction. Where there is a cathode reduction reaction, there may be different results depending on the respective material properties. A conductive wire appears when there is an anode oxidation and a cathode reaction. A salt bridge may be formed in a galvanic cell as a result of half cel l reactions. Galvanic corrosion happens when a salt bridge formed. Iron atoms lose electrons when corrosion happens, which form an electron transport circulation in the corrosion environment . Despite the basic requirements to form corrosion, oxygen content, water appearance, temperature, pH and other secondary factors a ffect corrosion results or even a ffect the formation of chemical corrosion products. In theory, the higher the moisture content , oxygen ap pearance, and temperature, the higher the corrosion rate will be in the system. 2.6 Corrosion Inhibitors To prevent severe corrosion happens to engineer materials on structural main bodies, which may result in catastrophes by reducing structure different kinds of corrosion inhibitors are used in civil engineering field . Higher demands for restraining corrosion on structural materials are pursued because of heavy pollution . There are two types of c orrosion inhibitors applied in nowadays engineering industry. The first one is ino rganic inhibitors and another is organic inhibitors. Among inorganic inhibitors, they

PAGE 24

11 are divided into four types. A nodic corrosion inhibitors, cathodic inhibitors, mixed inh ibitors which combine anodic inhibitors and cathodic inhibitors together. And the last one: volatile corrosion inhibitors (Myer 2005). Anodic inhibitors reduced the rate of anodic reactions. When oxidation reaction happens on anode s , these inhib it ors produ ce a n oxide film on the metal surface which needs to be protected. And these are called anodic inhibitors. Passivation happens during the test after corrosion sediments settle and the oxidized layer which is made of deposit s prevent s metal which needs to b e protected from being exposed to corrosion environment . Passivation makes oxygen a nd water become unavailable . Anodic inhibitors are usually called sacrifi cial anodes ( Sacrificial Anodes n.d) because they corrode first during corrosion exposure ( Cicek 201 4). For cathodic corrosion inhibitors, there are two ways to slow down the corrosion rate. In the first method, cathodic inhibitors decrease the cathodic reaction rate by using cathode poisons, and on the second way, selected deposits were generated on the cathodic areas . The appearance of deposits restrain s the rate of cathodic reactions. Anodic and cathodic corrosion inhibitors can be chose n into different tests and sometimes mixed inhibitors can reduce the rate of the anodic and cathodic reaction rate a t the same time more efficiently . 2.7 The Rules in Selection of Inhibitors When selecting corrosion inhibitors into tests, properties of the structural material which need to be pr otected became the first concern. M etal activity sheet became a dominant r ole in choosing inhibitors when different metals appear at the same time . The metal activity sheet shows the different activities undergo chemical reactions among types of metal . Comparisons on

PAGE 25

12 reactiveness among different metals can be made by using th e s heet. ( Figure 2 4). The metal is less active than the ones listed above it on the metal activity sheet. During a n electrochemical reaction process, whe n anodic inhibitors are chosen, t he part which is made of more reactive metal will work as the sacrificia l anode and it will be exposed to corrosion . Sacrificial anodes are widely used in boating system when sea water works as an electrolyte (Wankhede n.d) . And metal pieces which are made of zinc are chosen because zinc is more active than iron and it is econ omical to be used . The places where sacrificial anode attached decide the frequency of changing sacrificial anodes. Because the sacrificial metal is much more anodic and active, it will lose more electrons than the original metal bulk which needs to be pr otected. When sacrificial metal start s to release electrodes, oxidization reaction will initiate. With the carrying on of oxidization reaction, corrosion is accumulate d on the sacrificial anode. Ideally , corrosion happens only on sacrificial met al and the bulk metal should be well protected without being attacked by corrosion. Apart from properties of co nstruction materials, the problem how long the expected service life should be co nsidered to avoid over protecting or lacking protection. Methods and steps ensured before applying FRP sheets on. Besides, other factors may affect the formation of corrosion products like local environment effects (like humidification, p H and frozen and thaw, etc.) and , safety requirements should be reconsidered as well.

PAGE 26

13 2.8 Zinc Corrosion Inhibitors From the metal activity series, when st eel needs to be protected, almost all elements list beyond steel can work as corrosion inhibitors. Zinc can be selected as the sacrificial anode to protect steel product s in the following thesis testing because the elemen t zinc is more active than iron. Zi nc is listed in a higher position than iron . Besides, zinc inhibitors are economical to use. With the increasing of hydroxyl ions around the metal protected, the pro tected metal is passivated because of the products in electrochemical reactions pr event th e metal being exposed to corrosion . Zinc coatings and zinc spraying are much more efficient in prevent ing co rrosion accumulation than zinc coins or zinc blocks . There are two reasons for this . The first reason is that spraying zinc can function well as a t hin barrier, which make s water and oxygen un available to metal pieces . Secondly, zinc spraying is ful ly covered on surfaces of meta l pieces, which need to be protecte d. It will start to protect steel parts as soon as the electrochemical reaction starts. Pr otection becomes uniformly spread on steel members where zinc spray covers. To reveal the use of zinc inhibitors applied in the structural field, f our types of tests were conducted during the entire research . There were corrosion simulation tests on strips and beams, friction tests between steel strips and CFRP sheets after different time of corrosion exposure, tensile tests for steel strips, epoxy coupons and CFRP coupons . W hen comes to beam testing, bending tests were selected to crush beams and load capa city of each beam was tested. 2.9 Materials All experimental tests conducted were based on the properties of ASTM A992 steel, type CF130 CFRP sheet and epoxy coupons. A992 steel is widely used in beam constr uction in bridges and buildings, and it is mainl y used in producing I beams and H beams with a 345 MPa yield

PAGE 27

14 strength and a 200 GPa (30,000,000 psi) elastic modulus. According to dash sheet , MBrace CFRP (CF 130 ) has a n equivalent 0.165 mm thickness with a 240 GPa fiber modulus, which is re latively low when compared with other MBrace products like MBrace CF530, which has a 3800 MPa tensile strength and with the average thickness of 0.176 mm. Fiber reinforced polymer is a newly emer ged reinforcement material, and it can be produced in differ ent forms and FRP sheets can be applied on the surface of cleaned structural membe can increase the load bearing capacity of beams, slabs and other structural members. Two parts of epoxy adhesives are used to glue CFRP sh eet s with steel surfaces. Epoxy resin adhesives fall into the non Newtonian group ( Optimizing Viscosity for Epoxy Adhesives n.d) , which does with a low viscosity of 1600 cps at 75 ° F. Epoxy resin adhesives are best used in applications where the gap between two members is thin enough or the area that needs application i s relatively small .

PAGE 28

15 Figure 2 1 Chemical and Mechanical Properties of A992 Steel ( ASTM A992 Beam n.d) .

PAGE 29

16 Figure 2 2 Corrosion on steel bars . Figure 2 3 Corrosion on Brooklyn Bridge. ( https://commons.wikimedia.org/wiki/File:(Brooklyn_Bridge)_On_Corrosion,_Bridges_and_Sk ys crapers.jpg, 2017)

PAGE 30

17 Figure 2 4 Metal activity sheet ( The reactivity series of metals n.d

PAGE 31

18 CHAPTER III STEEL STRIPS TESTING 3.1 C o rrosion Simulation T ests for Steel S trips There are four prerequisites that help form co rrosion. During corrosion stimulation, both anode and cathode should exist at the same time. Secondly , electrolyte and the contact between cathode and anode ( must be electrically conductive ) should be fulfilled. T he current in the system shoul d return to t he original source, and it is called a current return path. The changing of potentials on anodes and cathodes may result in current start to flow in another direction. These are call ed All m aterials used i n tests should be prepared with caution because different processing procedures will affect the extent of corrosion sometimes even corrosion products . Steel and CFRP sheets are the main materials used in the present test . Steel strips with dimensions of 1 00 mm long 37 mm wide 3 mm thick were selected to use in steel strips testing. Two pieces of steel strips formed one test specimen . Before bonding CFRP sheets , the surface of each steel strip was cleaned by sand paper , a wire brush , and a grin der . Two re asons can explain why cleaning and polishing process es were necessary: firstly, rust existing on strips before wanted corrosion exposure may affect the final corr osion accumulation. Secondly, getting rid of rust before a corrosion simulation process begins makes a better bonding effect between steel substrate s and C FRP sheets. Type CF130 CFRP sheet s were chosen , which have an intermediate fiber modulus 2 27 GPa and a tensile st rength of 3 , 800 MPa ( MBRACE® CF 130 n.d). Usually a single layer of CFRP sheet is in a nominal thickness of 0.1 65

PAGE 32

19 a dimension of 100 mm 37 mm 0.165 mm were used to strengthen ste el strips, and they were glued in the middle of two steel strips. Steel strips were bonded on both si des. Zinc coins were selected as zinc inhibi tors in tests for steel strips; they are all 31.75 mm +/ 1.59 mm in d iameter and 2.54 mm in thickness. A blade saw machine was used here to cut zinc coins from zinc cast rods (all zinc cast rods came from rotome tal.com). The unit weight of cast rod used in the test is 3.84 lbs. / ft., and they are 99.6% min zinc. The e xact same type of zinc coins w as applied in beam testing. There are two parts of adhesives, which compose the epoxy resin used in the test. Epoxy is regarded as one kind of strong glue, and it is commonly used in mechanical bond ing . F iber reinforced polymer products ( laminates, sheet s , etc.) can be bonded to the surface of structural members , making the parts bond together tightly . The weight ratio of epoxy part A to part B is 3:1. Fine and quick mixing should ensure the quality of good gluing effect. E xcessive glue should be removed, and an equivalent thickness of epoxy adhesive about 1mm was required. It took 7 days for the epoxy to be fully cured at room temperature (20 ). The properties of the epoxy resin are listed in Figure 3 2 . A g as to rch was used to melt zinc coins . W elding was used to make zinc coins connect with steel strips without gaps , but it did not work well. That is why another method (using a gas torch) wa s selected. The melting ensured that connected to the test specimens, which l ed to better electrical contact with steel and zinc coins when e lectrochemical corrosion happened . Tapes and plastic ties were used . They were used to attach zinc coins with . During corrosion exposure, it is poss ible that zinc coins corrode badly or the corrosion is uneven on entire surface s of steel specimens . When this situation happens, ta pes and

PAGE 33

20 plastic bands will play key role s to prevent zinc coins from becoming surfaces. Chloride corrosion: a 3.5% concentration NaCl s olution was used during the whole corrosion simulation process. A normal concentration used in experimental tests is 3.5 %, and it is used commonly as chemical media in studying corrosion on metals. Na+ and Cl are both corpuscles, which are electrically conductive. The appearance of salt improve d the conductivity of water , and w ater was the electrolyte in the corrosion test. The tra nsfer of electron s and metal atoms were allowed because of the electrical conductivity of the electrolyte and the differences between electrical potential betw een anode and cathode. Electrical potential differences exist ing between dissimilar metals make s ure galvanic corrosion happens. Conductive w ires were used to form anode and cathode connection. Anode ( steel strips bonded with CFRP sheet) was connected to the positive side of the battery charger to make sure excessive corrosion happened to steel strips and the cathode linked to the negative side of the battery . The cathode can be any size of steel pieces and one 200 mm 37 mm 3 mm steel piece was selected. Same cleaning and polishing procedures were followed but on the cathode . The connection stimula te d producing rust on the steel strips bonded with CFRP surrounded by zinc coins. A b attery charger provided input current into the reaction containers. It was used to accelerate the electrochemical corrosion process. For the reason, the formation of corrosi on naturally is relatively slow and limited time allowed for a research conducts in the lab so that is why speeding up the corrosion is necessary. Voltage input on the strips boosts to generate the corrosion. The battery charger 1050 PE (charger 2/10/50 AM P) from Schumacher Electric was used in all corrosion stimulation tests at a current of 10 amp.

PAGE 34

21 The v olt amper e meter is used to monitor the volt amper e changing patterns when the corrosion propagating during the test. Ampere readings were measured every t wo hours at several random points on steel strips and the average value s of the ampere reading s were compared. Differences between the ampere reading s from starting the test and to the at end of the test can be treated as an index to show the corrosion inh ibitors Strain gauges were bonded to the CFRP sheet. The distance between the strain gauges was 15.875 mm from center to center and the first strain gauge located 19 mm from the center of the specimen . Strain gage reading s developed when the spec imen was loaded. With the tests carried , strain s increase d before the CFRP de bond ed and dropped after failure happened. Normally, there are two methods to evaluate the effect of corrosion inhibitor. The first one is measuring the mass differences between metal pieces before tests and metal pieces after corrosion exposure. They were labeled under MBT ( mass before corrosion exposure) and MAT (mass after corrosion exposure) in mass loss tables. Another method is using electrochemical measurements. Besides, m icroscopy can be used in a chemistry lab. In here, mass and area measurements were conducted to get mass loss curve and corrosion rate of each strip was calculated . With the corrosion generating in containers during 96 exposure , the load capacity of each steel strip was decreasi ng while the mass loss of steel strips and the mass loss of zinc coins were increasing. The mass of zinc was measured too for each category. 3.2 Tension Test Tensile tests were conducted on steel strips to abil ity working as corrosion inhibitors . From pictures taken during the whole corrosion exposure process , more and

PAGE 35

22 more rust was formed with time counting . After 144 hours corrosion exposure, steel strips without zinc protection were seriously damaged. S ome strips even cracked into two parts and they cannot be tested in tension tests. Steel strips with zinc coins looked better than steel strips without zinc coins only steel strips which were exposed in corrosion less th an 96 hours were tested to make sure the same number of specimens were tested in both groups. Two groups of steel strips were taken into research: Steel strips without corrosion inhibitors there are five categories steel strips after a different time of corrosion exposure (0 hours, 24 hours, 48 hours, 72 hours, and 96 hours respectively). The entire time length of corrosion exposure is 144 hours but heavy corrosion happened to steel strips w hich exposed in chloride corrosion more than 96 hours. Moreover, it became difficult to conduct tensile tests on these steel pie ces, which were severely damaged because it was difficult to attach corroded steel pieces in tensile MTS machine. That is why on ly strips without corrosion, strips which were exposed in chloride corrosion environment less than 96 hours were taken into consideration for comparison here. From the load capacities measured of steel strips without zinc coins , a decreasing curve of load capacity can be drawn. In the group s teel strips without zinc coins , the average ultimate load capacity of steel strips was 23.4 kN when there was no corrosion happened , and it dropped to 8.3 kN after 24 e generating of rus ts, the average load capacity dropped to 5.9 kN, 4.9 kN, 3.6 kN after 48 hours, 72 hours and 96

PAGE 36

23 respectively. C orrosion happened mu ch more intense on the first 24 hour corrosion exposure from the decreasing curve . The l oad capacity o f steel strips tested in tensile tests as well at different time categories and they are listed in T able 3 2. A similar decreasing trend line was witnessed during the entire corrosion tests, but by contrast, it showed some d ifferent results from steels strips without any protection: higher load capacities after corrosion tests when compared with strips without zinc coins. The failure load dropped from a same 23.4 kN to 8.7 kN after 24 hour chloride corrosion, a nd it decreased to an average value of 6.7 kN after 96 hours . Intense corrosion happened mostly at the first 24 hour corrosion exposure, but t he final 6.7 kN that was near twice the load capacity of strips without zinc coins When compared with t he group of steel strips without zinc protection, the average load capacity of steel strips with zinc coin shows zin c played a key role in preventing steel from a significant decr ease o f ultimate load capacity . It shows when zinc coins work as corrosion inhibitors in chloride cor rosion test, it d id some help in preventing rust accumulating. The c orrosion rate of each steel strip and the efficiency of inhibitors were analyzed. Mass loss o f strips without zinc corrosion and strips without zinc protection were measured and the average mass loss of each category was calculated. Mass loss of each specimen was weighed after cleaning . The curve which i s the summary of mass loss of each specime n i s provided together with the percentage of average mass in Figure 3 23 and Figure 3 24 . The mass of strips without zinc c oins decreased more, which was a 20.72 g in average than strips with

PAGE 37

24 protection (17.94 g in average) and this phenomenon complied with load capacity changing pattern. The more severe corrosion happened, the more mass loss was revealed. The summary of the mass of each coin that measured after a different time of corrosion exposure was listed in T able 3 5. When making a comparison betw een these two tables, significant differences of mass loss can be drawn. The average mass loss of steel strips in two groups became similar in general. between two groups became larger. For the group without zinc pr ot ection, the average mass loss was 6.76 g and the average mass loss of strips with zinc protection was 2.60g. A s imilar difference was found when compared with the average mass loss percentage. Zinc coins worked as sacrificial metal s and the drop of mass showed their anti corrosion effec t. Zinc corroded first during chloride corrosion because it behaved much more reactive than steel as a result of its chemical properties. So, when corrosion was prop a gating , zinc coins became vulnerable and their mass decre ased until the tests stopped. The c leaning process for zinc coins after corrosion before weighing became difficult because zinc coins became porous after being attacked by electrochemical corrosion. When the protection became less efficient, steel strips s tarted to corrode. On each strip bonded with CFRP sheets applied in the test, there are four zinc coins (two on each side) attached to the steel surface. The mass loss of every coins was measured and then the average value was calculated . When selecting z inc as the corrosion inhibitor in the te sts above, it is obvious that zinc did some effect s on protecting steel from rusting . The average corrosion rate of steel strips with zinc protection after 144 hour corrosion exposure was 11.594 c m/year and by contrast, the corrosion rate of steel strips without any corrosion protection was 14.342 cm /year. Significant differences of corrosion protection effect between unprotected steel and zinc protected steel can be detected in a long term.

PAGE 38

25 Standard deviation and coefficient of variation in math were used to evaluate the dispersion in a series of samples. Low standard devi ation means almost all data approach to the mean value. After conducting tension tests on strips , load capacities of strips are data point s used in dispersion analysis and the standard deviation was not low enough for the reason that data points are too limited to decide whether load capacities followed a standard deviation curve. The trend line was used to see the general changing pattern. Ba sed on corrosion inhibitor efficiency in different categories listed in Table 3 8 , it showed when zinc coins work as sacrificial anodes in corrosion exp but they still decreased the corrosion rate somehow when they appeared around steel parts . Better connection between zinc coins and steel strips surfaces, more zinc coins or other types of zinc in hibitors like spraying zinc can be chosen to provide better corrosion protection . 3.3 Friction T est The purpose of conducting fric tion tests on corroded steel strips was to reveal the anti corrosion effect of zinc coins not only from their appearance s b ut also from physical properties . The more time of corrosion exposure on steel strips, the higher coefficient of friction was tested during friction tests. From the appearance s of strips after being im merged in chloride corrosion environment for different time periods , strips beca me rougher as time went by. The c oefficient of friction represents the frictional characteristics of tested materials. Before calculating the friction coefficient , the weight of each CFRP sheet was measured at 0 hours, 24hours, 48hours, 72hours and 96 hours respectively and the summary of all masses were listed in T able 3 9 and T able 3 10 . Only four categories (0 hour, 24 hour, 48 hour and 72 hour) of steel strips were tested i n friction test because, steel strips corroded badly and only small pieces of them can be found corrosion exposure . Steel strips became shorter and limited surfaces area o f steel strips will be

PAGE 39

26 provided. As a consequence, the reading from friction test system was limited and became less accurate . A l inear relationship between time and velocity can be drawn and the acceleration value can be calculated because it is the slope of the curve. According to F friction = N norminal , the coefficient of friction was calculated. Average value of each was summarized. When performing the friction test, steel strips and CFRP coupons in each category were selected. A 50 mm diameter pulley s ystem was the device to test coefficient of friction. A small weight was used to make sure CFRP sheet would slide on steel surface s and sliding coefficient of friction could be tested . Steel strips were installed on a flat surface and a small weight of coi n was attached to CFRP coupons using a tape a nd a thread . What needs to be paid attention to is that the weight should be selected properly: it should not be so large that it may pull the CFRP while sliding happens , so there will be no reading on the software because of no contact between CFRP coupons and steel surface; On the other hand, if the weight is too small, it is likely that the weight cannot make CFRP coupons slide on the surface of steel strips, which may happen to steel strips that are heavily deteriorated. The weight applied to the friction test were 0.00298 k g and 0.00594 kg . When steel strips were not seriously damaged, which the time of corrosion exposure was less than 24 hours, the weight of 0.00298 kg was heavy enough to slide the CFRP surfaces. The mass of each CFRP coupons was weighed before friction tests started. From the friction coefficient table, friction coefficient increa sed with the propagating of rust on steel str ips, and when compared with steel strips without corrosion protection, the coefficient of friction was much lower on strips bonded with zinc coins. From the definition of friction coefficient, that means surfaces of steel

PAGE 40

27 strips with zin c protection were m uch smoother than unprotected steel strips after corrosion exposure . 3.4 Coupon Test Epoxy coupons and CFRP coupons were made and then they were exposed to chloride corrosion environment together with steel strips to illustrate how corrosion influence the ir loading capacities . T he dimension of each epoxy coup on was in a nominal length of 101.6 mm and a thickness of 5.08 mm. Epoxy resin was mixed and then poured in a mode, and they were ready to be exposed to chloride corrosion after 7 days curing time . T he thickness of CFRP coupons were 0.165 mm and they were cut into 203.2 mm long and 12.5 mm wide segments. As same as steel strips testing groups , different time categories were made. Chloride corrosion was s imulated on co upons for 0 hours, 24 hours, 48 hour s, 72 hours, 96 hours, 120 hours and 144 hours respectively. The load capacity of each coupon was tested in tension test and the average value was calculated afterward . Load capacity of each epoxy coupon and CFRP coupon is listed in T able 3 11 and T able 3 12. From T able 3 11 , the average load capacity of epoxy coupons decreased from 2.10 kN to a n average 2.04 sion exposure. The slight difference between these two average values means epoxy res in is not as vulnerable as steel member itself after chloride c orrosion. F or the reason that metal is m uch more electrical conductive, more electrochemical corrosion will happen on steel strips. F or load capacities of CFRP coupons, no much differences between the average load capacities of coupo ns before corrosion and coupons after 144 hours was found . And this scenario approved that CFRP sheets were corrosion resistant .

PAGE 41

28 Table 3 1 Tensile capacity of strips without zinc coins. With out Zinc ID Ultimate Load (kN) Av erage Ultimate Load (kN) Standard Deviation (kN) Coefficient of Variation 0 1 24.7 23.4 1.8 0.08 0 2 21.2 0 3 25.5 0 4 22.0 24 1 8.9 8.3 0.8 0.10 24 2 9.2 24 3 7.9 24 4 7.2 48 1 6.3 5.9 0.2 0. 03 48 2 5.8 48 3 5.7 48 4 5.9 72 1 4.6 4.9 0 .8 0. 16 72 2 5.7 72 3 3.7 72 4 5.4 96 1 3.7 3.6 1.8 0. 50 96 2 6.3 96 3 1.3 96 4 3.0 Coefficient of variation=standard deviation/average ultimate load 0 1: The first steel strip with out at 0 hour. x y: The yth steel strip with out

PAGE 42

29 Table 3 2 Tensile capacity of strips with zinc protection. With Zinc ID Ultimate Load (kN) Average Ultimate Load (kN) Standard Deviation (kN) Coefficient of Variation z 0 1 24.7 23.4 1.8 0.08 z 0 2 21.2 z 0 3 25.5 z 0 4 22.0 z 24 1 7.8 8.7 0.9 0.10 z 24 2 8.3 z 24 3 10.2 z 24 4 8.6 z 48 1 8.3 7.3 0.9 0.12 z 48 2 8.0 z 48 3 6.7 z 48 4 6.2 z 72 1 7.4 7.0 1.29 0.18 z 72 2 4.8 z 72 3 7.8 z 72 4 8.0 z 96 1 6.8 6.7 1.18 0.18 z 96 2 7.4 z 96 3 7.9 z 96 4 4.8 Coefficient of variation=standard deviation/average ultimate load z 0 1: The first steel strip with zinc c z x corrosion exposure.

PAGE 43

30 Table 3 3 Mass loss of strips without zinc coins. MBT: Mass before corrosion test. MAT: Mass after corrosion test. ML: Mass loss during the whole corrosion simulation process. Aver Mass Loss (%) =average mass loss/average mass before tests. ID MBT (g) MAT (g) ML (g) Mass Loss Av g (g) Av g Mass Loss ( % ) ID MBT (g) MAT (g) ML (g) Mass Loss Av g (g) Av g Mass loss ( % ) 0 198.9 198.9 0.0 0.00 0.0 24 1 199.2 198 .3 0.9 0.74 0.4 96 1 201.2 192.3 8.9 8.72 4.3 24 2 198.8 198.1 0.7 96 2 200.3 191.6 8.7 24 3 197.5 196.8 0.7 96 3 198.5 1 87.4 11.1 24 4 198.5 197 .9 0.6 96 4 197.6 188.9 8.7 24 5 201.2 199.4 0.8 96 5 198.6 192.4 6.2 48 1 199.8 197.4 2.4 2.64 1.3 120 1 203.6 191.7 11.9 11.38 5.7 48 2 198.1 196.0 2.1 120 2 202.5 191.8 10.7 48 3 202.3 199.1 3.2 120 3 197.6 1 88.4 9.2 48 4 198.7 196.2 2.5 120 4 197.5 184.3 13.2 48 5 199.1 196.1 3.0 120 5 197.9 186.1 11.9 72 1 197.7 190.8 6.9 6.76 3.4 144 1 199.5 177.0 22.5 20.72 10.3 72 2 199.5 192.3 7.2 144 2 197.9 179.5 18.4 72 3 201.6 194.2 7.4 144 3 198 .1 180.2 17.9 72 4 198.9 192.7 6.2 144 4 203.3 179.7 23.6 72 5 198.7 192.6 6.1 144 5 202.7 181.5 21.2

PAGE 44

31 Table 3 4 Mass loss of strips with zinc coins. MBT: Mass before corrosion test. MAT: Mass after corrosion test. ML: Mass loss during the whole corrosion simulation process. Avg Mass Loss (%) =average mass loss/average mass before tests. ID MBT (g) MAT (g) ML (g) Mass Loss Avg (g) Avg Mass Loss (%) ID MBT (g) MAT (g) ML (g) Mass Loss Avg (g) Avg Ma ss loss (%) 0 198.9 198.9 0.0 0.00 0.0 24 1 199.1 198.7 0.4 0.42 0.2 96 1 201.3 196.7 4.6 5.24 2.6 24 2 200.1 199.6 0.5 96 2 197.6 192.0 5.6 24 3 199.4 199.1 0.3 96 3 198.0 192.2 5.8 24 4 189.5 189.2 0.3 96 4 198.0 193.2 4.8 24 5 192.6 192.0 0.6 96 5 201.3 195.9 5.4 48 1 200.1 199.5 0.6 0.96 0.5 120 1 200.9 185.0 15.9 9.98 5.0 48 2 197.9 197.2 0.7 120 2 199.3 192.1 6.8 48 3 200.2 198.6 1.6 120 3 198.2 188.8 9.4 48 4 198.1 196.5 1.6 120 4 197.8 191.4 6.4 48 5 189.9 18 9.6 0.3 120 5 198.1 176.7 11.4 72 1 197.3 194.6 2.7 2.60 1.3 144 1 198.2 186.6 11.6 17.94 9.0 72 2 201.3 197.6 3.7 144 2 199.0 178.0 21.0 72 3 199.1 196.8 2.3 144 3 197.0 176.9 20.1 72 4 197.9 195.9 2.0 144 4 200.4 173.8 26.6 72 5 198. 8 196.5 2.3 144 5 200.1 189.7 10.4

PAGE 45

32 Table 3 5 Mass loss of zinc coins. ID MBT (g) MAT (g) ML (g) Avg Mass Loss (g) ML (%) ID MBT (g) MAT (g) ML (g) Avg Mass Loss (g) ML (%) 0 1 15.7 15.7 0 0 0 96 1 15.2 11.2 4.0 5.96 39.5 0 2 14.6 14.6 0 96 2 14.7 10.1 4.6 0 3 16.0 16.0 0 96 3 16.8 13.9 2.7 0 4 15.3 15.3 0 96 4 16.5 5.3 10.2 0 5 15.5 15.5 0 96 5 15.7 10.0 5.7 0 6 15.0 15.0 0 96 6 14.8 12.3 2.5 0 7 14.8 14.8 0 96 7 14.5 10.2 4.3 0 8 15.7 15.7 0 96 8 13.7 9.5 6.2 0 9 14.8 14.8 0 96 9 16.1 5.0 10.1 0 10 16.3 16.3 0 96 10 15.6 7.0 8.6 0 11 13.8 13.8 0 96 11 15.3 8.9 6.4 0 12 15.5 15.5 0 96 12 15.2 7.4 7.8 0 13 16.4 16.4 0 96 13 14.4 5.7 8.7 0 14 17.0 17.0 0 96 14 14.9 10.9 4.0 0 15 16 .0 16.0 0 96 15 15.3 10.0 5.3 0 16 15.5 15.5 0 96 16 13.9 6.6 7.3 0 17 13.5 13.5 0 96 17 14.4 10.0 4.4 0 18 12.5 12.5 0 96 18 14.8 10.2 4.6 0 19 13.0 13.0 0 96 19 14.8 7.5 7.3 0 20 10.5 10.5 0 96 20 15.2 10.7 4.5 24 1 15.7 13 2.7 1.67 11.0 120 1 15.1 11.7 3.4 6.86 45.4 24 2 14.6 13.2 1.4 120 2 14.9 6.2 8.7 24 3 16.0 14.5 1.5 120 3 16.8 8.0 8.8 24 4 15.3 13.9 1.4 120 4 16.5 10.2 6.3 24 5 15.5 13.8 1.5 120 5 15.2 7.1 8.1 24 6 15.0 13.2 1.8 120 6 14.8 8.9 5 .7 24 7 14.8 13.5 1.3 120 7 15.3 9.7 5.6 24 8 15.7 14.5 1.2 120 8 14.7 6.0 8.7 24 9 14.8 12.9 1.9 120 9 15.5 12.9 2.6 24 10 16.3 14.6 1.7 120 10 16.0 8.8 7.2 24 11 13.8 12.0 1.8 120 11 15.2 7.6 7.6 24 12 15.5 14 1.5 120 12 14. 6 6.1 8.5 24 13 16.4 14.6 1.8 120 13 13.5 5.5 8.0

PAGE 46

33 24 14 17.0 15.1 1.9 120 14 Note: Only 13 zinc coins were found after test. 7 coins corroded badly and cannot be found. 24 15 16.0 14.6 1.4 120 15 24 16 15.5 14.2 1.3 120 16 24 17 13 .5 12.0 1.5 120 17 24 18 12.5 10.9 1.6 120 18 24 19 13.0 11.2 1.8 120 19 24 20 10.5 8.2 2.3 120 20 48 1 13.0 10.1 2.9 2.52 16.7 144 1 14.7 6.6 8.1 8.96 59.4 48 2 14.2 12.0 2.2 144 2 14.7 5.0 9.7 48 3 15.5 13.1 2.4 144 3 15.3 4.7 10.6 48 4 17.8 14.8 3.0 144 4 16.2 7.0 9.2 48 5 16.5 14.1 2.4 144 5 15.2 6.5 8.7 48 6 14.0 11.4 2.6 144 6 14.6 7.2 9.4 48 7 13.4 11.2 2.2 144 7 15.2 4.1 11.1 48 8 13.2 11.2 2.0 144 8 15.1 5.8 9.3 48 9 14.0 11.4 2.6 1 44 9 15.8 7.5 8.3 48 10 16.5 14.7 1.8 144 10 14.2 6.2 6.0 48 11 16.5 13.8 2.7 144 11 14.5 7.3 8.2 48 12 17.2 14.9 2.3 Note: Only 11 zinc coins were found after test. 9 coins were missing. 48 13 13.0 10.4 2.6 48 14 14.8 12.2 2.6 48 15 17.0 14.1 2.9 48 16 18.5 15.8 2.7 48 17 15.3 12.6 2.7 48 18 16.5 14.1 2.4 48 19 15.0 11.2 3.8 48 20 13.5 12.0 1.5 72 1 15.2 12.1 3.1 4.44 29.4 72 2 16.1 12.0 4.1 72 3 13.8 6.9 6.9 72 4 16.2 14.2 2 .0 72 5 17.3 14.9 2.7 72 6 15.4 14.2 3.2 72 7 15.5 13.9 1.6 72 8 17.2 10.4 6.8 72 9 14.8 6.0 8.8 72 10 15.0 6.6 8.4 72 11 14.7 10.5 4.2 72 12 14.5 10.5 4.0 72 13 15.2 5.0 10.2 72 14 16.8 14.3 2.5

PAGE 47

34 72 15 13.2 11.7 1.5 72 16 15.1 9.8 5.3 72 17 14.4 10.2 4.2 72 18 14.7 12.3 2.4 72 19 15.4 10.9 4.5 72 20 13.9 11.5 2.4 MBT: Mass before corrosion test. MAT: Mass after corrosion test. ML: Mass loss due to corrosion . Avg Mass Loss (%) =average mass loss/average mass before tests*100%

PAGE 48

35 Table 3 6 Corrosion rate of strips without zinc coins. (Based on ASTM G1 Standard Practice for Preparing, Cleaning, and Evaluation Corrosion Test Specimens). T (h) Corrosion Rate (cm /year) Avg Corrosion Rate ( cm /year) Mass Loss (g) A ( c m²) 0 0.000 0.000 0.0 162 .220 24 2.619 2.153 0.9 159.782 24 2.036 0.7 159.846 24 2.037 0.7 159.772 24 1.746 0.6 159.738 24 2.327 0.8 159.884 48 3.563 3.921 2.4 156.604 48 3.097 2.1 157.662 48 4.736 3.2 157.068 48 3.705 2.5 156.882 48 4.506 3.0 154.784 72 7.039 6.960 6.9 151.924 72 7.364 7.2 151.544 72 7.589 7.4 151.13 0 72 6.347 6.2 151.396 72 6.460 6.1 146.346 96 6.915 6.812 8.9 149.602 96 6.825 8.7 148.166 96 8.663 11.1 148.944 96 6.819 8.7 148.306 96 4.838 6.2 148.964 120 8.924 8.823 11.9 124.006 120 8.087 10.7 123.044 120 8.181 9.2 104.572 120 9.963 13.2 123.206 120 8.959 11.9 123.524 144 15.905 14.342 22.5 1 09.626 144 12.818 18.4 111.238 144 11.641 17.9 119.166 144 17.513 23.6 104.432 144 13.836 21.2 118.744 T: Time of exposure in hours; Corrosion Rate= (K*W)/ (A*T*D) K: A constant used in calculating corrosion rate. K = 87 , 600 in here. W: Mass loss in grams;

PAGE 49

36 A: Surface area of strips in cm 2 ; Measure the average length, width and thickness of each steel strip. A=2*(length*width) +2*(length*thickness) +2*(width*thickness). D: Density of the material (steel). For steel used in the tests, D = 7.85 g/cm 3 .

PAGE 50

37 Table 3 7 Corrosion rate of strips with zinc under chloride corrosion. (Based on ASTM Standard Practice for Preparing, Cleaning, and Evaluation Corrosion Test Specimens). T (h) Corrosion Rate (c m/ye ar) Avg Corrosion Rate (c m/year) W (g) A (c m²) 0 0.000 0.000 0.0 162.220 24 1.164 1.222 0.4 159.838 24 1.455 0.5 159.764 24 0.872 0.3 159.91 0 24 0.873 0.3 159.806 24 1.745 0.6 159.848 48 0.885 1.436 0.6 157.624 48 1.046 0.7 155.564 48 2 .425 1.6 153.384 48 2.385 1.6 155.982 48 0.440 0.3 158.448 72 2.671 2.583 2.7 156.646 72 3.668 3.7 156.342 72 2.288 2.3 155.824 72 1.982 2.0 156.426 72 2.305 2.3 154.63 0 96 3.578 4.067 4.6 149.428 96 4.371 5.6 148.922 96 4.444 5.8 151.72 4 96 3.722 4.8 149.902 96 4.221 5.4 148.724 120 11.330 6.763 15.9 130.502 120 4.254 6.8 148.652 120 6.304 9.4 138.662 120 4.035 6.4 147.516 120 7.893 11.4 134.304 144 5.862 11.594 1 1.6 153.344 144 13.980 21.0 116.408 144 13.265 20.1 117.4 26 144 18.861 26.6 109.29 0 144 6.003 10.4 134.264 T: Time of exposure in hours; Corrosion Rate= (K*W)/ (A*T*D) K: A constant used in calculating corrosion rate. K= 87 , 600 in here. W: Mass loss in grams;

PAGE 51

38 A: Surface area of strips in cm 2 ; Measure the av erage length, width and thickness of each steel strip. A=2*(length*width) +2*(length*thickness) +2*(width*thickness). D: Density of the material (steel). For steel used in the tests, D = 7.85 g/cm 3 = 0.00785 g/mm 3 . Table 3 8 In hibitor efficiency of zinc coins. ID Average Corrosion Rate ( c m /year) ID Average Corrosion Rate ( c m/year) Inhibitor Efficiency (%) 24z 1.22 24n 2.15 43.24 48z 1.44 48n 3.92 63.38 72z 2.58 72n 6.96 62.89 96z 4.07 96n 6.81 40.30 120z 6.77 120n 8.82 2 3.35 144z 11.60 144n 14.32 19.16 24z, 48z, 72z, 96z, 120z, 144z: ID for steel strips with zinc protection in different category. 24n, 48n, 72n, 96n, 120n, 144n: ID for steel strips without zinc protection in each different category . Inhibitor Effi ciency ( % ) = 100*(CR unprotected CR protected )/ CR unprotected

PAGE 52

39 Table 3 9 Friction test between CFRP sheet and steel strips surface without zinc. T (hours) Weight (kg) M cfrp (kg) Acceleration (m/s²) F (N) Fricti on Coefficient Avg Friction Coefficient 0 0.00298 0.0055 1.5145 0.004513 0.31 0.22 0 0.00298 0.0055 2.2825 0.006802 0.19 0 0.00298 0.0055 2.3225 0.006921 0.18 0 0.00298 0.0055 2.1638 0.006448 0.21 0 0.00298 0.0055 2.2548 0.006719 0.19 24 0.002 98 0.0059 1.3608 0.004055 0.30 0.29 24 0.00298 0.0059 1.3702 0.004083 0.30 24 0.00298 0.0059 1.4885 0.004436 0.28 24 0.00298 0.0059 1.5646 0.004663 0.27 24 0.00298 0.0059 1.5709 0.004681 0.27 48 0.00594 0.0062 1.5542 0.009232 0.65 0.68 48 0.00594 0.0062 1.2552 0.007455 0.71 48 0.00594 0.0062 1.5499 0.009206 0.66 48 0.00594 0.0062 1.5846 0.009413 0.65 48 0.00594 0.0062 1.2661 0.007521 0.71 72 0.00594 0.0055 2.2734 0.013504 0.61 0.76 72 0.00594 0.0055 1.3129 0.007799 0.81 72 0.00594 0.005 5 1.3399 0.007959 0.80 72 0.00594 0.0055 1.5274 0.009073 0.76 72 0.00594 0.0055 1.3485 0.008007 0.80 T: time. Weight: weight of coins applied to cause the acceleration. M cfrp : weight of CFRP sheet. Acceleration: Time velocity linear curve can be draw n from the machine and acceleration is the slope of it. F: normal force applied in the system. F = Weight * Acceleration Friction coefficient = ((Weight *g) (Weight + Mcfrp) * Acceleration))/ (Mcfrp * g) g = 9.81 N/kg.

PAGE 53

40 Table 3 10 Friction test between CFRP sheet and steel strips surface with zinc coins. T (hours) Weight (kg) M cfrp (kg) Acceleration (m/s²) F (N) Friction Coefficient Avg Friction Coefficient 0 0.00298 0.0055 2.3549 0.007018 0.18 0.21 0 0.002 98 0.0055 2.5276 0.007532 0.15 0 0.00298 0.0055 2.1293 0.006345 0.21 0 0.00298 0.0055 1.9786 0.005896 0.24 0 0.00298 0.0055 1.8793 0.005600 0.25 24 0.00298 0.0059 1.6551 0.004932 0.26 0.24 24 0.00298 0.0059 1.9541 0.005823 0.21 24 0.00298 0.0059 1.7562 0.005233 0.24 24 0.00298 0.0059 1.7249 0.005140 0.25 24 0.00298 0.0059 1.8589 0.005540 0.23 48 0.00594 0.0062 1.8034 0.010712 0.60 0.61 48 0.00594 0.0062 1.6213 0.009631 0.64 48 0.00594 0.0062 1.7903 0.010634 0.61 48 0.00594 0.0062 1.743 7 0.010358 0.62 48 0.00594 0.0062 1.8231 0.010829 0.60 72 0.00594 0.0055 1.7201 0.010217 0.72 0.72 72 0.00594 0.0055 1.6805 0.009982 0.73 72 0.00594 0.0055 1.7997 0.010690 0.71 72 0.00594 0.0055 1.8532 0.011008 0.69 72 0.00594 0.0055 1.6204 0.00 9625 0.74 T : time. Weight: weight of hollow coins applied to cause the acceleration. Mcfrp: weight of CFRP sheet. Acceleration: Time velocity linear curve can be drawn from the machine and acceleration is the slope of it. F: normal force applied in the system. F = Weight * Acceleration Friction coefficient = ((Weight * g ) (Weight + Mcfrp) * Acceleration))/ (Mcfrp * g) G = 9.81 N /kg.

PAGE 54

41 Table 3 11 Tensile capacity of epoxy coupons (exposed to corrosion environment up to 144 h ours). ID Failure Load (kN) Tensile Strength (MPa) Average Failure Capacity (kN/[MPa]) Standard Deviation (kN/[MPa]) Coefficient Of Variation 0 1 2.0 24.3 2.10 [25.2] 0.47 [5.75] 0.22 0 2 2.9 34.8 0 3 1.8 21.4 0 4 1.7 20.2 24 1 1.1 1 3.5 1.80 [21.5] 0.80 [9.23] 0.44 24 2 3.1 36.6 24 3 1.8 21.5 24 4 1.2 14.5 48 1 2.9 34.6 3.15 [37.5] 0.54 [6.48] 0.17 48 2 3.5 42.0 48 3 2.4 28.4 48 4 3.8 45.0 72 1 1.8 21.9 2.18 [26.1] 0.54 [5.83] 0.25 72 2 3.0 35.0 72 3 1.6 20.0 72 4 2.2 27.5 96 1 2.0 23.6 1.9 [22.3] 0.12 [1.43] 0.06 96 2 1.7 20.0 96 3 2.0 23.3 96 4 1.9 22.2 120 1 1.7 19.9 1.7 [19.9] 0.31 [3.46] 0.18 120 2 2.0 23.3 120 3 1.9 22.1 120 4 1.2 14.3 144 1 1.3 15.1 2.04 [24.0 ] 0.48 [5.96] 0.24 144 2 2.3 27.7 144 3 2.6 31.0 144 4 2.0 23.3 C ross section area =12.7 mm (width) × 5.08mm (thickness) = 64.516 mm 2

PAGE 55

42 Table 3 12 Tensile capacity of CFRP coupons (exposed to corrosion environmen t up to 144 hours). ID Failure Load (kN) Tensile Strength (MPa) Average Failure Capacity (kN/[MPa]) Standard Deviation (kN/[MPa]) Coefficient Of Variation 0 1 8.8 4166.7 5.9 [2793.6] 1.7 [797.2] 0.29 0 2 4.9 2320.1 0 3 5.2 2462.1 0 4 4.7 2 225.4 24 1 4.9 2320.1 4.8 [2260.9] 0.7 [327.0] 0.14 24 2 3.9 1846.6 24 3 5.8 2746.2 24 4 4.5 2130.7 48 1 4.7 2225.4 4.8 [2272.7] 1.4 [644.0] 0.28 48 2 3.1 1467.8 48 3 4.5 2130.7 48 4 6.9 3267.0 72 1 5.5 2604.9 4.5 [2130.7] 1. 3 [636.1] 0.30 72 2 6.0 2840.9 72 3 3.9 1846.6 72 4 2.6 1231.1 96 1 5.4 2556.8 5.0 [2379.3] 0.2 107.8 0.05 96 2 5.0 2367.4 96 3 4.8 2272.7 96 4 4.9 2320.1 120 1 3.9 1846.6 4.8 [2260.9] 2.8 [1336.4] 0.59 120 2 2.9 1373.1 120 3 2.7 1278.4 120 4 9.6 4545.5 144 1 3.7 1751.9 5.2 [2450.3] 1.7 [825.1] 0.34 144 2 3.3 1562.5 144 3 6.2 2935.6 144 4 7.5 3551.1 Cross section area= 12.7 mm (width) × 0.165 mm (thickness) = 2.0955 mm 2

PAGE 56

43 (a) (b) Figure 3 1 Epoxy adhesive: (a) Masterbrace SAT 4500 Part A; (b) Masterbrace SAT 4500 Part B. Figure 3 2 Properties of epoxy.

PAGE 57

44 Figure 3 3 Battery charger used in tests. Figure 3 4 Schematic of test coupon.

PAGE 58

45 (a) (b ) (c) (d) (e) Figure 3 5 Corrosion test for steel strips wit h zinc coins :(a) steel strips at 24 hr before test; (b) 48 hr; (c) 72 hr ; (d) steel strips at 96 hr before test ; (e) steel strips at 120 hr before test.

PAGE 59

46 (a) (b) (c) ( d) (e) (f) Figure 3 6 Corrosion in progress: (a) 24 hr; (b) 48 hr; (c) 72 hr; (d) 96 hr; (e) 1 20 hr; (f) 144 hr.

PAGE 60

47 (a) (b) (c) (d) (e) (f) Figure 3 7 Corrosion process completed: (a) 24 hr; (b) 48 hr; (c) 72 hr; (d) 96 hr; (e) 120 hr; (f) 144 hr.

PAGE 61

48 (a) (b) (c) (d) ( e) Figure 3 8 Zinc coins: (a) 0 hr; (b) 24 hr; (c) 48 hr; (d) 72 hr; (e) 96 hr.

PAGE 62

49 Figure 3 9 Mass loss of zinc coins. (a) (b) Figure 3 10 Tension test setup for strips with CFRP using MTS machine: (a) before loading (for w/o zinc at 0 hours with strain gauge); (b) after failure (for w zinc at 48 hours without strain gauge). 0 2 4 6 8 10 12 0 50 100 150 200 Mass loss (g) Time (hours)

PAGE 63

50 (a) (b) (c) (d) (e) (f) Figure 3 11 Strips with zinc and strips without zinc after corrosion: (a) 0 hr ; (b) 24 hr ;(c) 48 hr ; (d) 72 hr ; (e) 96 hr. (f) all strips.

PAGE 64

51 Figure 3 13 Load CFRP strain of strip without zinc at 0 hrs. 0 5 10 15 20 25 0 0.01 0.02 0.03 Load (kN) Strain #1 #2 #3 Figure 3 12 Accelerate corrosion.

PAGE 65

52 Figure 3 14 Load CFRP strain of strip without zinc at 24 hrs. Figure 3 15 Load CFRP strain of strip without zinc at 4 8 hrs. 0 1 2 3 4 5 6 7 8 9 10 0 0.0005 0.001 0.0015 0.002 Load (kN) Strain #1 #2 #3 0 1 2 3 4 5 6 0 0.0005 0.001 Load (kN) Strain #1 #2 #3

PAGE 66

53 Figure 3 16 Load CFRP strain of strip without zinc at 72 hrs. Figure 3 17 Load CFRP strain of strip without zinc at 96 hrs. 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 0.000 0.001 0.001 0.002 0.002 Load/ kN Strain #1 #2 #3 0 0.5 1 1.5 2 2.5 3 3.5 0 0.0005 0.001 0.0015 0.002 Load (kN) Strain #1 #2 #3

PAGE 67

54 Figure 3 18 Load CFRP strain of strip with zinc at 0 hrs. Figure 3 19 Load CFRP strain of strip with zinc at 24 hrs. 0 5 10 15 20 25 30 0 0.002 0.004 0.006 0.008 0.01 Load (kN) Strain #1 #2 #3 0 1 2 3 4 5 6 7 8 9 0 0.001 0.002 0.003 0.004 Load( kN) Strain #1 #2 #3

PAGE 68

55 Figure 3 20 Load CFRP strain of strip with zinc at 48 hrs. Figure 3 21 L oad CFRP strain of strip with zinc at 72 hrs. 0 1 2 3 4 5 6 7 0 0.0005 0.001 0.0015 0.002 0.0025 Load (kN) Strain #1 #2 #3 0 1 2 3 4 5 6 7 0 0.0005 0.001 0.0015 Load( KN) Strain #1 #2 #3

PAGE 69

56 Figure 3 22 Load CFRP strain of strip with zinc at 96 hr s . 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0 0.0005 0.001 0.0015 0.002 Load (kN) Strain #1 #2 #3

PAGE 70

57 (a) (b) Figure 3 23 Strips without zinc mass loss pattern: (a) strips mass loss in grams; (b) mass loss percentage. (Mass loss percentage= (mass loss/original mass) * 100%) 0 5 10 15 20 25 30 0 50 100 150 200 Mass loss (g) Time (hours) 24 48 72 96 120 144 0 2 4 6 8 10 12 0 24 48 72 96 120 144 Mass loss (%) Time (hours)

PAGE 71

58 (a) ( b) Figure 3 24 Strips with zinc mass loss pattern: (a) strips mass loss in grams; (b) mass loss percentage (Mass loss percentage= (mass loss/original mass) * 100%). 0 5 10 15 20 25 30 0 50 100 150 200 Mass loss (g) Time (hours) 24 48 72 96 120 144 0 2 4 6 8 10 12 0 24 48 72 96 120 144 Mass loss (%) Time (hours)

PAGE 72

59 Figure 3 25 Corrosion rate of steel strips without zinc coins. Figure 3 26 Corrosion rate of steel strips with zinc coins. 0 5 10 15 20 0 50 100 150 Corrosion rate/(cm/year) Time/(h) 0 5 10 15 20 0 50 100 150 Corrosion rate/(cm/year) Time/(h)

PAGE 73

60 Figure 3 27 Comparison between the average ultimate load of strips without and with zinc protectio n. Figure 3 28 Load displacement of strips without zinc. 0 5 10 15 20 25 0 20 40 60 80 Load (kN) Time (hours) steel strips without zinc steel strips with zinc 0 5 10 15 20 25 30 0 0.05 0.1 0.15 0.2 Load ( k N) DISP laser (mm) 0-1 24-1 48-1 72-1 96-1

PAGE 74

61 Figure 3 29 Load displacement of strips with zinc coins. Figure 3 30 Schematic of friction test. 0 5 10 15 20 25 30 0 0.05 0.1 0.15 0.2 Load (kN) DISP laser (mm) z-0-1 z-24-1 z-48-1 z-72-1 z-96-1

PAGE 75

62 F igure 3 31 Friction test. Figure 3 32 Friction coefficient between CFRP sheet and steel without zinc protection after corrosion. 0 0.2 0.4 0.6 0.8 1 1.2 0 20 40 60 80 100 Friction coefficient Time (hours)

PAGE 76

63 Figure 3 33 Friction coefficient between CFRP sheet and steel with zinc after corrosion. 0 0.2 0.4 0.6 0.8 1 1.2 0 20 40 60 80 100 Friction coefficient T ime (hours)

PAGE 77

64 (a) (b) Figure 3 34 Coupons: (a) Epoxy; (b) CFRP.

PAGE 78

65 (a) (b) (c) (d) (e) (f) Figure 3 35 Epoxy coupons after conditioning: (a) 0hr. (b)24hr;(c) 48hr; (d)72hr; (e) 96hr; (f)120hr.

PAGE 79

66 (a) (b) Figure 3 36 Tension test: (a) epoxy coupon at failure; (b) laser extensometer. (a) (b) Figure 3 37 Tension test for CFRP coupons: (a) before tension. (b) after tension.

PAGE 80

67 (a) (b) (c) (d) (e) (f) (g) Figure 3 38 Epoxy coupons after tension test: (a) 0 hr; (b) 24 hr; (c) 48 hr; (d) 72 hr ; (e) 96 hr; (f) 120 hr; (g) 144 hr.

PAGE 81

68 (a) (b) (c) (d) (e) (f) (g) Figure 3 39 CFRP coupons after tension test: (a) 0 hr; (b) 24 hr; (c) 48 hr; (d) 72 hr ; (e) 96 hr; (f) 120 hr; (g) 144 hr.

PAGE 82

69 CHAPTER IV BEAM TEST The chloride corrosion simulation acce lerate d the process of accumulating rust . After being exposed to corrosion, the purpose of bending tests on beams is to illustrate the de bonding effect between CFRP sheet s and after corrosion s. Beams in different categories were prepared, deteriorated and th en tested under flexural tests. Among structural members, beams take the responsibility of carrying loads together with columns and other structural elements. Beams carry transverse loading that means despite vertical loadings, end moments are carried by beams too . And among beams categorization, girders are the most vital beams i n carrying loads, and b ending is the main way to deflect a beam . As an important member in structural parts, the significance of preventing beams from corrosion is addressed. A total of nineteen steel beams bonded with CFRP sheets were taken into research and they were processed in three groups : zinc coins and beams with zinc anti rusting spraying . There was one beam being tested as control beam and it was the beam without corrosion (no corrosion exposure) . Other 18 beams are divided in three different corrosion exposure time: beams after 48 hours corrosion exposure, 96 corrosion exposure, and 144 expos ure respectively. 4 .1 Corrosion Simulation A992 structural steel is normally used in I bea m construction . B eams were prepared, applied corrosion and tested in bending tests . Wide flange I beam W 100 x 19 was cut by using a blade saw into 1000 mm long beam segments . A slot was created on the one side on the flange by using the b lade and the cutting depth was 3 0 m m from bottom flange . After being cut by the saw,

PAGE 83

70 beams were cleaned and polished using a grinder to make sure no existing rusts. Then CFRP wor ked as a s trengthening material to improve load capacities if beams under bending tests. Same type of CFRP sheet s were used for strengthening the bending capacity of steel beams. CFRP sheet was cut into 700 mm ( L) 100 mm ( W) 0.165 mm ( T ) segments . Epoxy adh es ives were applied to ensure good bonding effect. Zinc coins were cu t from 31.75 mm +/ 1.59 mm in diameter and 2.54 mm in thickness from zinc cast rods . There were 18 zinc coins placed on bottom flange of each beam. Except beams bonded with zinc coins, a nother type of corrosion inhibitors being used in beam testing was spraying zinc . 3M TM 16 501 Zinc Spray was selected in painting because of its high purity which was made of 97% pure zinc. T he zinc spraying depth was 97 mm from the bottom flange (tension area ). Firm and direct connection between zinc coins and steel surfaces is one of the necessary to prevent corrosion of the steel. Tapes and plastic ties were used to attach zinc coins with the surface of bea ms. Similar chloride corrosion s imulation proce dures were continued as what had done in steel strips testing. Two bea ms were tested in each category. Every two beams which experienced same preparation were placed in a container and two timber made blocks were placed underneath beam s to make sure beams were not contact the bottom of container directly. To illustrate the corrosion charger and three pieces of steel strips in a dimension of 100 mm long 37 mm w ide 3 mm connected to the negative side of the battery and were immerged in Nacl solution . The depth of immersion was 40 mm on each container . Two battery chargers provided i nput 2A ampere on each beam to start corrosion.

PAGE 84

71 4 .2 Flexure Test One point load ing on middle span was applied gradually by a 20 kip MTS machine. The maximum displacement set for loading machine was 200 mm. Two point supports were provided under the bottom flange of steel beams (which surface was bonded with CFRP sheet and would exper ience tension ) in all bending tests and each support was located 50 mm away from the edge. Six strain gauges on each beam were organ ized and attached to the tension side of each beam on CFRP sheet for str ain analysis while loading ( see F igure 4 1 ) . Load st rain curves were achieved after dealing with strain data. In theor y, the maximum strain would happen near middle span that means strain gauges #3 or #4 (F igure 4 1) normally will tell the maximum strain because the steel fibers near middle span will experi ence the maximum elongation while tensile force applies . And e ither strain gauge #1 or #6 will experience the least strain among six strain gauges. From testing results ( F i gure 4 34 to F igure 4 42), the peak value of strain reading s on #3 or #4 strain gau ges located on the right of readings from other strain gauges. And this showed the small fiber particles attached near middle span experienced the largest elongation. Pi gauge s are P i shaped displacement transducers. The length of each gauge is l00mm and two gauges were used near the mid span in two locations and the y were used to test the strain of steel beam in a horizontal direction during the test. The pi ga u ge (labeled as #3) which was applied on top of steel beams experience d compression because it was installed in compression si de while another one experienced tension (Pi gauge #5). When one beam experiences bending, the length of #3 gauge became smaller while the length of #5 became larger. A l oad cell was installed beyond tested beams on the mid dle span before bending tests conducted. And as a transducer, it record ed the load that beam s experienced in seconds s in flexural tests. A p lacement potentiometer was selected to be used in the test to record

PAGE 85

72 displacem ent changing pattern of beam s in the vertical direction. A l oad displacement curve of each beam was listed in Figure 4 15 to F igure 4 24 . Based on testing results, beams without any corrosion protection dropped from 58 kN to an average value of 52.45 kN after 48 hour corrosion exposu re and it dropped sharply to an average 46.45 kN and 40.05 kN after 96 hour and 144 hour corrosion exposure respectively . High standard deviation shows the data is widely spread. Like beams without corr osion inhibitors, Load capacities of beams with zinc coins decreased with the deterioration of beams. After 144 hour corrosion exposure, the average load capacity of beams with zinc coins was 44 .10 kN, which was higher than 40 .05 kN which was the average reading of beams without protection . It showed zinc co ins protected beams from rusting effectively to some extent. From T able 4 3 , it showed a slightly decreasing pattern of load capacities on beams with spraying zinc protection. After 48 hour corrosion exposure with an average ultimate load 52.00 kN and ther e was only less than 2 kN difference between this group and the average ultimate capacity of beams which after 144 hour corrosion exposure. Mass of beams before corrosion happened and after corrosion happened was weighed. They are summarized in T able 4 4 to Table 4 6 and they were listed under the label MBT. From tables, mass loss of beams with zinc coins and beams with zin c spray after corrosion exposure is much smaller than beams without any protection. It showed when zinc worked as inhibitors on steel m embers, it provides some helps to prevent heavy weight loss.

PAGE 86

73 Table 4 1 Load capacities of beams without any corrosion protection. Time of Corrosion Exposure ( hour ) ID Ultimate Load (kN) Av g Ultimate Load (kN) Standard Deviation (kN) Coefficient of Variation 0 C ontro l Beam 58.0 58.00 0.00 0.00 48 BWOP1 51.4 52.45 1.05 0.02 48 BWOP2 53.5 96 BWOP3 48.1 46.45 1.65 0.04 96 BWOP4 44.8 144 BWOP5 45.3 40.05 5.15 0.13 144 BWOP6 35.0 BWOP: Beams without corros ion protection (no corrosion inhibitors). Two beams were tested in each category. Table 4 2 Load capacities of beams with zinc coins. Time of Corrosion Exposure ( hour ) ID Ultimate Load (kN) Av g Ultimate Load (kN) Standar d Deviation (kN) Coefficient of Variation 0 Control Beam 58.0 58.00 0.00 0.00 48 BWZC1 50.0 50.70 0.70 0.01 48 BWZC2 51.4 96 BWZC3 48.3 49.15 0.85 0.02 96 BWZC4 50.0 144 BWZC5 41.6 44.10 2.50 0.06 144 BWZC6 46.6 BWZC: Beams with zinc co ins. Table 4 3 Load capacities of beams with zinc spray. Time of Corrosion Exposure ( hour) ID Ultimate Load (kN) Av g Ultimate Load (kN) Standard Deviation (kN) Coefficient of Variation 0 Control Beam 58.0 58.00 0.00 0.00 48 BWZS1 50.5 52.00 1.50 0.03 48 BWZS2 53.5 96 BWZS3 55.4 54.55 0.85 0.02 96 BWZS4 53.7 144 BWZS5 48.6 50.65 2.05 0.04 144 BWZS6 52.7 BWZS: Beams with spraying zinc.

PAGE 87

74 Table 4 4 Mass loss of beams without cor rosion protection. BWOP: Beams without corrosion protection (no corrosion inhibitors). Two beams are tested in each category. MBT: Mass before corrosion test. MAT: Mass after corrosion test. ML: Ma ss loss during the whole corrosion simulation process. Avg Mass Loss = (Mass loss average/average MBT) * 100% Table 4 5 Mass loss of beams corrosion with zinc coins. BWZC: Be ams with corrosion inhibitors protection. ( Zinc coins from cast rod). Two beams are tested in each category. MBT: Mass before corrosion test. MAT: Mass after corrosion test. ML: Mass loss during the whole corrosion simulation process. Avg Mass Loss = (Mass loss average/average MBT) * 100% Time of Corrosi o n Exposure (hour) ID MBT (kg) MAT (kg) ML (kg) Mass Loss Aveg (kg) Avg Mass Loss ( % ) 48 BWOP1 19.14 19.14 0.00 0.01 0.05 48 BWOP2 19.00 18.98 0.02 96 BWOP3 19.12 19.10 0.02 0.02 0.11 96 BWOP4 18.90 18.8 8 0.02 144 BWOP5 19.06 19.04 0.02 0.03 0.16 144 BWOP6 19.18 19.14 0.04 Time of Corrosion Exposure (hour) ID MBT (g) MAT (g) ML (g) Mass Loss Avg (g) Avg Mass Loss ( % ) 48 BWZC1 19.08 19.08 0.00 0.01 0.05 48 BWZC2 19.24 19.22 0.02 96 BWZC3 19.16 19.14 0.02 0.02 0.10 96 BWZC4 19.20 19.18 0.02 144 BWZC5 19.26 19.26 0.00 0.02 0.10 144 BWZC6 19.12 19.08 0.04

PAGE 88

75 Table 4 6 Mass loss of beams corrosion with zinc spray. BWZS: Beams with corrosion inhibitors protection. ( Zinc spray from cast rod). T wo beams are tested in each category. MBT: Mass before corrosion test. MAT: Mass after corrosion test. ML: Mass loss during the whole corrosion simulation process. Avg Mass Loss = (Mass loss average/average MBT) * 100% Time of Corrosion Exposure (hour) ID MBT (kg) MAT (kg) ML (kg) Mass Loss Avg (kg) Avg Mass Loss ( % ) 48 BWZS1 19.12 19.10 0. 02 0.01 0.00 48 BWZS2 19.36 19.36 0.00 96 BWZS3 19.14 19.12 0.02 0.02 0.10 96 BWZS4 19.12 19.10 0.02 144 BWZS5 19.14 19.12 0.02 0.02 0.10 144 BWZS6 19.10 19.08 0.02

PAGE 89

76 Table 4 7 Corrosion rate of beams without zinc coins. (Based on ASTM G1 Standard Practice for Preparing, Cleaning, and Evaluation Corrosion Test Specimens). K: constant = 87600 in calculating corrosion rate. D: Density of the material (steel) = 7.85 g/c m³ T: Time of exposure in hours; Corrosion Rate = (K * W) / (A * T * D) W: Mass loss in grams; A: Surface area of beams in cm 2 ; Measure the average length, width and thickness of each steel beam. A = 2 *(length * width) + 2 *(length * thickness) + 2 *(width * thickness). Time (h) Corrosion Rate (mm/year) Avg Corrosion Rate (mm/year) W ( g) Avg W (g) A ( mm² ) Avg A ( mm² ) 48 0.000 1.913 0 10 607400 607800 48 3.825 20 608200 96 3.845 3.844 20 20 604600 604800 96 3.843 20 605000 144 2.577 3.864 20 30 601500 601650 144 5.151 40 601800 Figure 4 1 Beam bonded with CFRP and strain gauges.

PAGE 90

77 Figure 4 2 Beam with zinc coins and depth of immersion. Figure 4 3 Zinc Spray depth.

PAGE 91

78 Figure 4 4 Be am preparation. (a) (b) (c ) Figure 4 5 Setup for accelerated corrosion : (a) No corrosion; (b) after 24 hr corrosion exposure; (c)after 48 hr corrosion exposure.

PAGE 92

79 (a) (b) (c) (d) (e) (f) Figure 4 6 Beams after 48 hr corrosion exposure: (a ) ( b) Beams without protection; (c ) ( d) Beams with zinc coins ; (e ) ( f): Beams with zinc spray.

PAGE 93

80 (a) (b) (c) (d) (e) (f) Figure 4 7 Beams after 96 hr corrosion exposure: (a ) ( b) Beams without protection; (c ) ( d) Beams with zinc coins ; (e ) ( f): Beam s with zinc spray .

PAGE 94

81 (a) (b) (b) (d) (e) (f) Figure 4 8 Beams after 144 hr corrosion exposure: (a ) ( b) Beams without protection; (c ) ( d) Beams with zinc coins; (e ) ( f): Beams with zinc spray.

PAGE 95

82 Figure 4 9 Beams without any protection before corrosion simulation. (a) (b) Figure 4 10 Beams with zinc coins before corrosion : (a) zinc coins. (b) overview.

PAGE 96

83 Figure 4 11 Be ams with zinc spray protection before corrosion simulation. Figure 4 12 Beam deterioration.

PAGE 97

84 Figure 4 14 Beam failure. Figure 4 13 Test setup: (a) load cell and potentiometer; (b) PI gages. (a) (b)

PAGE 98

85 Figure 4 15 C ontrol beam. Figure 4 16 Beams orrosion exposure. 0 10 20 30 40 50 60 0 2 4 6 8 10 Load (kN) D isplacement (mm) Control Beam 0 10 20 30 40 50 60 0 2 4 6 8 10 Load (kN) D isplacement (mm) BWOP1 BWOP2

PAGE 99

86 Figure 4 17 Figure 4 18 0 10 20 30 40 50 60 0 2 4 6 8 10 Load (kN) Displacement (mm) BWZC1 BWZC2 0 10 20 30 40 50 60 0 2 4 6 8 10 Load (kN) Displacement (mm) BWZS1 BWZS2

PAGE 100

87 Figure 4 19 Figure 4 20 Beams with zinc coins after 96 0 10 20 30 40 50 60 0 2 4 6 8 10 Load (kN) D isplacement (mm) BWOP3 BWOP4 0 10 20 30 40 50 60 0 2 4 6 8 10 Load (kN) Displacement (mm) BWZC3 BWZC4

PAGE 101

88 Figure 4 21 Figure 4 22 0 10 20 30 40 50 60 0 2 4 6 8 10 Load (kN) Displacement(mm) BWZS3 BWZS4 0 10 20 30 40 50 60 0 2 4 6 8 10 Load (kN) Displacement(mm) BWOP5 BWOP6

PAGE 102

89 Figure 4 23 on exposure. Figure 4 24 0 10 20 30 40 50 60 0 2 4 6 8 10 Load (kN) Displacement(mm) BWZC5 BWZC6 0 10 20 30 40 50 60 0 2 4 6 8 10 Load (kN) Displacement(mm) BWZS5 BWZS6

PAGE 103

90 (a) (b) Figure 4 25 Load Pi strain for beams without protection at 48 hrs: (a) BWOP 1; ( b ) BWOP2. 0 10 20 30 40 50 60 0.000 0.005 0.010 0.015 0.020 Load (kN) Strain #3 #5 0 10 20 30 40 50 60 0.000 0.005 0.010 0.015 0.020 Load (kN) Strain #3 #5

PAGE 104

91 (a) ( b ) Figure 4 26 Load Pi strain for beams without protection at 96 hrs: (a) BWOP3; (b) BWOP4. 0 10 20 30 40 50 60 0.000 0.005 0.010 0.015 0.020 Load (kN) Strain #3 #5 0 10 20 30 40 50 60 0.000 0.005 0.010 0.015 0.020 Load (kN) S train #3 #5

PAGE 105

92 ( a ) (b) Figure 4 27 Load Pi strain for beams without protection at 144 hrs: (a) BWOP5; (b) BWOP6. 0 10 20 30 40 50 60 0.000 0.002 0.004 0.006 0.008 Load (kN) Strain #3 #5 0 10 20 30 40 50 60 0.000 0.001 0.002 0.003 0.004 Load (kN) S train #3 #5

PAGE 106

93 (a) ( b ) Figure 4 28 Load Pi strain for beams with zinc coins at 48 hrs: (a) BWZC1; (b) BWZC2. 0 10 20 30 40 50 60 0.000 0.005 0.010 0.015 0.020 Load (kN) Strain #3 #5 0 10 20 30 40 50 60 0.000 0.005 0.010 0.015 0.020 Load (kN) S train #3 #5

PAGE 107

94 ( a ) ( b ) Figure 4 29 Load Pi strain for beams with zinc coins at 96 hrs: (a) BWZC3; (b) BWZC4. 0 10 20 30 40 50 60 0.000 0.010 0.020 0.030 0.040 Load (kN) Strain #3 #5 0 10 20 30 40 50 60 0.000 0.001 0.002 0.003 0.004 Load (kN) Strain #3 #5

PAGE 108

95 ( a ) (b) Figure 4 30 Load Pi strain for beams with zinc coins at 144 hrs: (a) BWZC5; (b) BWZC6. 0 10 20 30 40 50 60 0 0.001 0.002 0.003 0.004 Load (kN) Strain #3 #5 0 10 20 30 40 50 60 0 0.001 0.002 0.003 0.004 Load (kN) Strain #3 #5

PAGE 109

96 (a) ( b ) Figure 4 31 Load Pi strain for beams with zinc spray at 48 hrs: (a ) BWZS1; (b) BWZS2. 0 10 20 30 40 50 60 0.000 0.005 0.010 0.015 0.020 Load (kN) Strain #3 #5 0 10 20 30 40 50 60 -0.001 0.001 0.003 0.005 0.007 Load (kN) Strain #3 #5

PAGE 110

97 ( a ) (b) Figure 4 32 Load Pi strain for beams with zinc spray at 96 hrs: (a) BWZS3; (b) BWZS4. 0 10 20 30 40 50 60 0 0.0005 0.001 0.0015 0.002 Load (kN) S train #3 #5 0 10 20 30 40 50 60 0 0.001 0.002 0.003 0.004 Load (kN) Strain #3 #5

PAGE 111

98 (a) (b) Figure 4 33 Load Pi strain for beams with zinc spray at 144 hrs: (a) BWZS5; (b) BWZS6. 0 10 20 30 40 50 60 0.000 0.005 0.010 0.015 0.020 Load (kN) Strain #3 #5 0 10 20 30 40 50 60 0.000 0.005 0.010 0.015 0.020 Load (kN) Strain #3 #5

PAGE 112

99 Figure 4 34 Load CFRP strain for beams without protection at 48 hrs. 0 10 20 30 40 50 60 0.000 0.002 0.004 0.006 0.008 Load (kN) S train SG1 SG2 SG3 0 10 20 30 40 50 60 0.000 0.002 0.004 0.006 0.008 Load (kN) Strain SG4 SG5 SG6

PAGE 113

100 Figure 4 35 Load CFRP strain for beams without protection at 96 hrs. 0 10 20 30 40 50 60 0 0.002 0.004 0.006 0.008 Load (kN) S train SG1 SG2 SG3 0 10 20 30 40 50 60 0 0.002 0.004 0.006 0.008 Load (kN) S train SG4 SG5 SG6

PAGE 114

101 Figure 4 36 Load CFRP strain for beams without protection at 144 hrs. 0 10 20 30 40 50 60 0.000 0.002 0.004 0.006 0.008 Load (kN) Strain SG1 SG2 SG3 0 10 20 30 40 50 60 0.000 0.002 0.004 0.006 0.008 Load (kN) Strain SG4 SG5 SG6

PAGE 115

102 Figure 4 37 Load CFRP strain for beams with zinc coins at 48 hrs. 0 20 40 60 0 0.0005 0.001 0.0015 0.002 Load (kN) Strain SG1 SG2 SG3 0 10 20 30 40 50 60 0 0.0005 0.001 0.0015 0.002 Load (kN) Strain SG4 SG5 SG6

PAGE 116

103 Figure 4 38 Load CFRP strain for beams with zinc coins at 96 hrs. 0 10 20 30 40 50 60 0 0.0001 0.0002 0.0003 0.0004 0.0005 0.0006 Load (kN) Strain SG1 SG2 SG3 0 10 20 30 40 50 60 0 0.0001 0.0002 0.0003 0.0004 0.0005 0.0006 Load (kN) Strain SG4 SG5 SG6

PAGE 117

104 Figure 4 39 Load CFRP strain for beams with zinc coins at 144 hrs. 0 10 20 30 40 50 60 0 0.001 0.002 0.003 0.004 Load (kN) Strain SG1 SG2 SG3 0 10 20 30 40 50 60 0 0.002 0.004 0.006 0.008 0.01 Load (kN) Strain SG4 SG5 SG6

PAGE 118

105 Figure 4 40 Load CFRP strain for beams with zinc spray at 48 hrs. 0 10 20 30 40 50 60 0 0.002 0.004 0.006 0.008 Load (kN) Strain SG1 SG2 SG3 0 10 20 30 40 50 60 0 0.002 0.004 0.006 0.008 Load (kN) Strain SG4 SG5 SG6

PAGE 119

106 Figure 4 41 Load CFRP for beams with zinc spray at 96 hrs. 0 10 20 30 40 50 60 0.000 0.002 0.004 0.006 0.008 Load (kN) Strain SG1 SG2 SG3 0 10 20 30 40 50 60 0.000 0.002 0.004 0.006 0.008 Load (kN) Strain SG4 SG5 SG6

PAGE 120

107 Figure 4 42 Load CFRP for beams with zinc spray at 144 hrs. 0 20 40 60 80 0 0.002 0.004 0.006 0.008 Load (kN) Strain SG1 SG2 SG3 0 20 40 60 80 0 0.002 0.004 0.006 0.008 Load (kN) Strain SG4 SG5 SG6

PAGE 121

108 CHAPTER V SUMMARY AND CONCLUSI ONS 5.1 Summary and Conclusions Two types of zinc inhibitors were used in the research. One was cut from zinc cast rod and another was in the form o f spraying zinc . According to tests results obtained , effect s of zinc inhibitors applied in structural steel bonded with CFRP sheets can be confirmed . There are four effect: (1) load capacities of steel mem bers without zinc tection or steel members with protection; (2) mass loss of zinc coins; (3) mass loss of steel members after diffe rent time of corrosion exposure; and (4) corrosion rate calculated according to cor rosion rate equation in AST M G1. A ll these factors illustrate how zinc influences the accumulation of rust on steel. From flexural tests for beams, z inc spray functioned much better than zinc protection , and acities. In the group in which b eams were protected by spraying zinc, minimal difference s of load capacities were found. 5.2 Recommendations for Future Research For future research aimed at t est ing the effect or efficiency of specific corrosion inhibitor s , stronger electrical contact among corrosion inhibitors and metal bulk, which needs protection should be ensured. In tests conducted in this research, gas torch , plastic tie s were use d to ensure electrical tra nsfer between cathode and anode, but they were not perfect bonding methods. Small gaps between z inc coins and steel surfaces were witnessed and s ome zinc coins lost contact with steel strips after hours of corrosion exposure because the bonding area wa s s eriously deteriorated . Moreover, the contact su rface should be as smooth as possible.

PAGE 122

109 When evaluating inhibitors effect, mass loss method and corrosion rate were selected to as indices in this thesis . Except for these two m ethods , e lectrochemical measurements like polarization curve ( Polarization Tec hniques n.d) method and the impedance measurement should be applied if permitted. The mass loss method used in pred icting corrosion rate is simple, and it is easy to manipulate in a labo ratory environment. However, mass loss methods require long time corro sion exposure, and the result could be biased if time of corrosion exposure is not long enough. Electrochemical measurements will give more detailed information about how inhibitors work during corrosion in a more accurate chemistry analysis . For beam te sting, more than one control beams should be tested if possible. And the average of load capacities of co ntrol beams will provide much more precise data . Similarly, when standard deviation and COV are used in probabilities and statics, they are often based on large numbe rs of samples. More data points and larger sample size will result in better assessment result s . Another consideration that should be given attention is how to evaluate the corrosion and the corrosion rate. During the tests mentioned above, how to determine the steel mass after corrosion is a problem because the cleaning process is m anually controlled and the base metal may lose more weight after cleaning . I t makes a difference between mass loss in corrosion exposure and mass mea sured on a s cale . When metal pieces are seriously damaged by corrosion, it becomes a challenge to remove only the rust.

PAGE 123

110 REFERENCES (2011) . West Conshohocken (PA), American Society for Testing and Materials. The Metals Compan y, "Corrosion Inhibitors." Corrosion of Linings & Coatings Corrosion Technology (2006): 55 68. (Jul. 17, 2017). Cicek, V. (2014). Corrosion engineering. Wiley, Hoboken (New Jersey). Cicek, V.(2011). "Cost of Corrosion and Use of Corrosion In hibitors." Corrosion Chemistry (2011): 39 41. reinforced polymer (FRP) Developments in Fiber Reinforced Polymer (FRP) Composites for Civil Engineering , 3 8. l Cooling Systems | GE Water , (Jul. 19, 2017). High performance and specialty fibers: concepts, technology and modern applications o f man made fibers for the future . (2016). Springer, Tokyo. Hollaway, L. C., and Teng, J. G. (2008). Strengthening and rehabilitation of civil infrastructures using fibre reinforced polymer (FRP) composites . CRC Press, Boca Raton.

PAGE 124

111 "MBRACE® CF 130 Unidirecti onal High Strength Carbon Fiber Fabric for the MBrace® Composit e Strengthening System." (n.d.). < http://www.bestmaterials.com/PDF_Files/MBRACE CF130_PDS.pdf > ( Jul. 17, 2017). Myer, K . (2005). Handbook of Environmental Degradation of Materials. William An drew Publishing. 81 82, 229 243. Activity of Metals , (Jul. 17, 2017). "Market Review s ." Market Overview American Composites Manufacturers Association (AC MA). American Composites Manufacturers Association (ACMA). GCSE Science: chemistry reactivity series, (Jul. 1 1, 2017). American Galvanizers Association , (Jul. 10 , 2017). strengt Construction and Building Materials , 90, 99 109. M, Hansson C. "An Introduction to Corrosion of Engineering Materials." Knovel . Meyers, A . "Cathode and Anode Half Cell Reactions." Study.com . Study.com. (Jul. 10 , 2017).

PAGE 125

112 Optimizing Viscosity for Epoxy Adhesives, Potting Compounds and Sealants | MasterBond.com, (Jul. 19, 2017). reinforced polymer (FRP) Strengthening and Rehabilitation of Civil Infrastructures Using Fibre Reinforced Polymer (FRP) Composites, 215 234. Polarization Techniques (n.d.): n. pag. Web. < http://nptel.ac.in/courses/113108051/module2/lecture10.pdf >(Jul.11,2017) Wankhede,A.(2017) Marine Insight , (Jul. 17, 2017).