See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/278244840

Sealing Small Movement Bridge Expansion Joints

Technical Report · June 2006

CITATIONS READS
4 2,187

3 authors, including:

Ramesh B. Malla
University of Connecticut
132 PUBLICATIONS   627 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

IDEA Rail Safety Project, RS-25 View project

All content following this page was uploaded by Ramesh B. Malla on 15 July 2015.

The user has requested enhancement of the downloaded file.

 SEALING OF
SMALL MOVEMENT BRIDGE EXPANSION JOINTS
Dr. Ramesh B. Malla, PI
Dr. Montgomery T. Shaw, Co-PI
Mr. Matu R. Shrestha, Graduate Research Assistant
and Ms. Smita Boob, Graduate Research Assistant
Prepared for
The New England Transportation Consortium
June 29, 2006
NETCR- 58 Project No. 02-6

This report, prepared in cooperation with the New England Transportation Consortium, does not
constitute a standard, specification, or regulation. The contents of this report reflect the views of
the authors who are responsible for the facts and the accuracy of the data presented herein. The
contents do not necessarily reflect the views of the New England Transportation Consortium or
the Federal Highway Administration.
 Technical Report Documentation Page
2. Government Accession No. 3. Recipient’s Catalog No.
.
1. Report No
N/A N/A
NETCR 58
4. Title and Subtitle 5. Report Date

June 29, 2006

Sealing of Small Movement Bridge Expansion Joints 6. Performing Organization Code

N/A
7. Author(s) 8. Performing Organization Report No.

Dr. Ramesh B. Malla, Dr. Montgomery T. Shaw, Mr. Matu R. NETCR 58

Shrestha, and Ms. Smita Boob
9. Performing Organization Name and Address 10 Work Unit No. (TRAIS)

University of Connecticut N/A

Department of Civil and Environmental Engineering
11. Contract or Grant No.

261 Glenbrook Road, Storrs, CT 06269-2037 N/A

13. Type of Report and Period Covered
12. Sponsoring Agency Name and Address

New England Transportation Consortium

FINAL REPORT
C/o Advanced Technology & Manufacturing Center
University of Massachusetts Dartmouth
151 Martine Street, Fall River, MA 02723
14. Sponsoring Agency Code

NETC 02-6. A study conducted in

cooperation with the U.S. DOT
15 Supplementary Notes

N/A
16. Abstract
Sealing of bridge expansion joint systems is important to protect the structural components below the joint from damage
due to water, salt, and other roadway debris. A new elastomeric foam-type joint sealant has been developed for sealing
small-movement bridge expansion joints. Laboratory tests including tension, compression, shear, bonding, salt water
immersion, temperature sensitivity, compression recovery, creep, stress relaxation, cure rate, tack time, and water
tightness were performed on this newly developed sealant to assess its mechanical and material characteristics. In
addition, loading-unloading behavior in tension and compression and effects of exposure to outdoor condition for 6.5
months period on its engineering properties were investigated. Similar tests were also performed on a currently available
commercial bridge joint sealant material for comparison purposes. The new silicone foam sealant showed an increase in
volume by about 70 % after the mixing of sealant components. The test results indicated lower stiffness, greater
extensibility, and better bonding associated with the foam sealant compared to the commercial sealant. Foam sealant
showed more resistant to fatigue with tensile deformation cycles and its stress relaxation rate was greater than that of
commercial sealant. The tack and cure time for foam sealant were small and no leakage was observed through the sealant
and joint interface. The sealant also did not exhibit any physical deterioration during prolonged exposure to natural
weathering elements; however it appeared to stiffen which might be due to oxidation and continuous sealant cure.
17. Key Words 18. Distribution Statement
Bridges, expansion joints, sealing, elastomer, No restrictions. This document is available to the public through the
foam, tests, stress, strain, tension, bonding National Technical Information Service, Springfield, Virginia 22161.

19. Security Classif. (of this report) 20. Security Classif. (of this page) 21. No. of Pages 22. Price

Unclassified Unclassified 112 N/A

Form DOT F 1700.7 (8-72) Reproduction of completed page authorized

ii
iii
 ACKNOWLEDGEMENTS

The research work reported herein was performed under Project NETC 02-6, sponsored by
New England Transportation Consortium. Special appreciation is expressed to the Technical
Committee members of this NETC project for their cooperation and considerable efforts in
providing us with their invaluable comments as well as highly useful information on various
joint sealing systems currently being used by their respective Transportation Agencies in New
England region. Our special gratitude goes to Mr. Robert Fura, the Chair of the NETC Technical
Committee, for his continuous support throughout the project duration. The supports and
facilities received from Department of Civil & Environmental Engineering and Institute of
Materials Science at University of Connecticut during the entire research period are gratefully
acknowledged. The support from Watson Bowman Acme Corporation, Amherst, New York,
who provided the solid sealant material for this study is greatly appreciated. Authors are also
thankful to all of those highway and transportation agencies who responded to the inquiries
relevant to this study.

iv
 TABLE OF CONTENTS

Page
Title Page …………………………………………………………..……………. i
Technical Report Documentation……………………………………………….. ii
Metric Conversion Chart………………………………………………………… iii
Acknowledgements……………………………………………………………… iv
Table of Contents………………………………………………………………… v
List of Figures…………………………………………………………………..... vii
List of Tables…………………………………………………………………….. ix

1. Introduction……………………………………………………........................ 1
1.1 Background………………………………………………………………. 1
1.2 Research Motivation……………………………………………………… 4
1.3 Project Objectives………………………………………………………… 4

2. Review of Bridge Expansion Joints…………………………………………. 6

2.1 Bridge Joint General Design Criteria………………………………….... 6
2.2 Types of Bridge Joints and Their Advantages and Disadvantages……… 8
2.2.1 Small Movement Joints…………………………………………… 8
2.2.1.1 Butt Joint………………………………………………… 9
2.2.1.2 Sliding Plate Joint……………………………………….. 9
2.2.1.3 Compression Seal Joint…………………………………. 11
2.2.1.4 Asphaltic Plug Joint…………………………………….. 13
2.2.1.5 Poured Sealed Joint……………………………………… 16
2.2.2 Medium Movement Joints………………………………………… 17
2.2.2.1 Strip Seal………………………………………………… 17
2.2.2.2 Finger Plate Joint……………………………………… 20
2.2.3 Large Movement Joints………………………………………….. 21
2.2.3.1 Plank Seal………………………………………………. 21
2.2.3.2 Modular Joints…………………………………………. 22
2.3 Expansion Joint in New England States and Their Features………….. 24
2.4 Evaluation of Performance of Expansion Joints ………………..……… 25
2.4.1 Common Joint Defects ……………………………………………. 25
2.4.2 Factors Influencing Joint Performance……………………………. 26
2.4.3 Bridge Deck Joint Evaluation……………………………………… 27

3. Development of New Sealant Material……………………………………. 32

3.1 Selection of Joint Sealant Material……………………………………… 32
3.2 Laboratory Development of Silicone Foam Sealant…………………… 34

4. Laboratory Evaluation Test Methodology……………………................. 37

4.1 Tension Test……………………………………………………………. 37
4.1.1 Loading Test ……………………………………………………. 37
4.1.2 Loading and Unloading Test…………………………………… 40
4.2 Compression Test………………………………………………………. 41

v
 4.2.1 Loading Test……………….…………………………………….. 41
4.2.2 Loading and Unloading Test…………………………………….. 42
4.3 Shear Test………………………………………………………………. 42
4.4 Salt Water Immersion Test……………………………………………... 44
4.5 Bond (Oven Aged) Test………………………………………………… 45
4.6 Stress Relaxation Test………………………………………………….. 47
4.7 Creep Test………………………………………………………………. 48
4.8 Compression Recovery Test……………………………………………. 49
4.9 Temperature Sensitivity Test…………………………………………… 50
4.9.1 High and Low Temperature Effects………………………………. 50
4.9.2 Freeze and Thaw Cycle Test………………………………………. 51
4.10 Cure Rate Test…………………………………………………………. 51
4.11 Tack Free Time Test ………………………………………………….. 52
4.12 Water Tightness Test………………………………………………….. 52
4.13 Tests on Weathered Sealant…………………………………………… 53
5. Laboratory Test Results and Discussion……….………………………… 55
5.1 Tension Test Results……………………………………………………. 55
5.1.1 Loading Test Results…………………………………….………... 55
5.1.2 Loading and Unloading Test Results…………………………….. 57
5.2 Compression Test Results………………………………………………. 59
5.2.1 Loading Test Results……………………………………………… 59
5.2.2 Loading and Unloading Test Results............................................... 59
5.3 Shear Test Results………………………………………………………. 61
5.4 Salt Water Immersion Test Results……………………………………… 62
5.5 Bond (Oven Aged) Test Results………………………………………… 66
5.6 Stress Relaxation Test Results………………………………………….. 68
5.7 Creep Test Results………………………………………………………. 73
5.8 Compression Recovery Test Results…………………………………… 73
5. 9 Temperature Sensitivity Test Results…………………………………… 74
5.9.1 High and Low Temperature Effects Test Results……………….... 74
5.9.2 Freeze and Thaw Cycle Test Results……………………………… 79
5.10 Cure Rate Test Results………………………………………………… 81
5.11 Tack Free Time and Water Tightness Test Results……………………. 82
5.12 Results of Tests on Weathered Sealants………………………………. 83
5.12.1 Tension Test Result………………………………………………. 85
5.12.2 Shear Test Results……………………………………………….. 87
5.12.3 Stress Relaxation Test Results…………………………………… 87
6. Conclusions ………………..………………………………………………… 90
7. Recommendations for Field Demonstration and Monitoring (Phase II) … 93
7.1 Recommendation to Proceed to Phase II ………………………………. 93
7.2 Recommended Major Tasks For Phase II ……………………………… 94
7.2.1 Pre-Field Installation Tasks ……………………………………… 94
7.2.2 Field Installation Tasks ………………………………………….. 96
7.2.3 Post-Field Installation and Monitoring Tasks……………………. 97

8. References……………………………………………………………………… 99

vi
 LIST OF FIGURES

Figure 1: Butt joint……………………………………………………….. 9

Figure 2: Sliding plate joint………………………………………………. 10
Figure 3: Compression seal………………………………………………. 12
Figure 4: Compression seal leakage………………………………………. 13
Figure 5: Asphalitc plug joint…………………………………………….. 14
Figure 6: Material flowing on to adjacent pavements……………………. 15
Figure 7: Section of failed joint…………………………………………… 15
Figure 8: Polymer poured-in-place seal………………………………….. 16
Figure 9: Debonding of silicone sealant………………………………… 17
Figure 10: Strip seal………………………………………………………. 18
Figure 11: Tearing of seal due to incompressible debris…………………. 19
Figure 12: Gland missing………………………………………………… 19
Figure 13: Finger Joint:(a) Plan view……..……………………………… 20
(b) Section A-A………………………………………………. 20
Figure 14: Plank seal……………………………………………………. 21
Figure 15: Damaged plank seal…………………………………………... 22
Figure 16: Modular joint…………………………………………………. 23
Figure 17: Damaged modular joint……………………………………… 23
Figure 18: Sample evaluation form……………………………………… 28
Figure 19: Summary of transportation departments’ evaluation of
bridge deck joints………………………………………………… 29
Figure 20: Schematic representation of foaming reaction………………… 35
Figure 21: Two silicone seal materials: (a) Silicone foam sealant……….. 36
(b) solid sealant………………………………………………… 36
Figure 22: Tension test specimen…………………………………………. 38
Figure 23: Instron testing machine (Model 1011) with tension specimen… 40
Figure 24: Tension test of sealant using Instron machine………………… 41
Figure 25: Compression of sealant………………………………………… 42
Figure 26: Shear test specimen..…………………………………………… 43
Figure 27: Shear test of foam sealant using Instron machine……………… 44
Figure 28: Diagram illustrating compression set………………………….. 50
Figure 29: Tensile behavior of foam sealant and solid sealant…………… 56
Figure 30: Tension loading and unloading………………………………… 58
Figure 31: Compression behavior of sealants…………………………….. 60
Figure 32: Compression loading and unloading………………………….. 60
Figure 33: Simple shear test results……………………………………… 61
Figure 34: Effects of salt water immersion on foam sealant….…………. 62
Figure 35: Effects of salt water immersion on solid sealant…..…………. 63
Figure 36: Comparison of effects of salt water on foam sealant
and solid sealant……………………………………………… 63
Figure 37: Fatigue of sealants with tensile deformation after oven aging
and cooling cycles. Curves are simplified C-T equation…….. 67
Figure 38: Bond Test: Stress-Strain at 5th extension……………………. 67

vii
Figure 39: Isothermal stress relaxation in tension for foam and solid.
Curves are KWW equation…………………………………… 71
Figure 40: Isothermal stress relaxation in compression for foam and solid.
Curves are KWW equation…………………………………. 72
Figure 41: Creep deformation of foam and solid in tension……………… 73
Figure 42: Compression Recovery of sealants…………………………… 74
Figure 43: Stress-Strain plots of samples after
high temperature conditioning………………………………… 76
Figure 44: Stress-Strain plots of samples after
cold temperature conditioning………………………………… 76
Figure 45: Effects of high and low temperature exposure
on sealant modulus…………………………………………. … 78
Figure 46: Effects of high and low temperature exposure
on sealant extensibility………………………………………… 78
Figure 47: Temperature variation in a single cycle……………………….. 80
Figure 48: Tension test results of sealants exposed to alternate
cold and room temperature…….………………………………. 80
Figure 49: Curing rate of foam and solid….………………………………. 82
Figure 50: Maximum and minimum temperature distribution….…………. 83
Figure 51: Cumulative precipitation…….………………………………… 84
Figure 52: Cumulative snowfall……….………………………………….. 84
Figure 53: Tension test results: weathered sealants……….………………. 86
Figure 54: Tension loading and unloading of weathered sealants…………. 86
Figure 55: Shear test results of weathered sealants……….………………… 87
Figure 56: Isothermal stress relaxation in tension.
Weathered foam sealant………………………………………… 89
Figure 57: Isothermal stress relaxation in tension.
Weathered solid sealant…………………………………………. 89

viii
 LIST OF TABLES

Table 1: Various Bridge Deck Joints in Practice in New England……… 24

Table 2: Joint Performance Ranking Based on Deterioration Rate…….. 30
Table 3: Sealant Components for Tension Test Specimen………………… 39
Table 4: Summary of Stress, Strain and Failure Mode Results
from Tension Tests…………………………………………… 57
Table 5: Results of Shear Test…………………………………………. 62
Table 6: Stress, Strain and Failure Mode Results from
Salt Water Immersion Test………………………………….. 65
Table 7: Salt Water Immersion Effects on Sealant Modulus
and Extensibility……………………………………………… 65
Table 8: Stress at 100 % Strain under Repeated Extensions
from Bond Test after Oven Aging…………………………… 66
Table 9: KWW Equation Parameters for Stress Relaxation Test…….... 69
Table 10: Stress, Strain and Failure Mode Results from
Temperature Sensitivity Test………………………………… 75
Table 11: Temperature Exposure Effects on
Sealant Modulus and Extensibility…………………………… 79
Table 12: Stress, Strain and Failure Mode Results from
Freeze and Thaw Cycle Test…………………………………. 81
Table 13 Curing Rate of Sealants………………………………………. 82
Table 14: KWW Equation Parameters for Stress Relaxation Test:
Weathered sealants……………………………………………. 88

ix
 1. INTRODUCTION

1.1 BACKGROUND

Bridges are continually moving. The movements could be due to temperature and

humidity changes; creep, shrinkage and cyclic effects; axial and flexural strains arising from

dead and live loads and prestressing; dynamic load effects including impact, vehicle braking

and lurching, centrifugal forces; and long-term movements such as those caused by

settlement and earth pressure (Lee, 1994). For the bridge to function as intended, it must be

capable of accommodating all such movements. The total combined motion of a bridge is

deceptively slow; however, forces involved are tremendous. For a 80-feet simple span steel

bridge with W36 x 210 longitudinal stringers, the force generated at one bearing, should its

movement be impeded because of a 40ºF temperature change, would be over 466 kips from

the beam alone (Mumber, 1993). So it is clearly evident that providing a means to

accommodate movements is extremely important to prevent the development of secondary

stresses in the bridge superstructure.

Movement is usually accommodated by bridge bearings and deck expansion joint

system; however, because bridge bearings are often sources of trouble, many bridges are

designed so that the entire structure takes care of movement without bearings. This is done

by making bridge piers flexible allowing for movements in the abutments and being sure the

approach pavement is not so rigid that the bridge is not allowed to expand and contract as

temperature dictate. When these methods cannot be used, however, expansion joint details

must be provided to allow the bridge free movement (NCHRP, 1977). A properly functioning

bridge deck joint accommodates the horizontal and vertical movements of the structure while

providing smooth ridability, low noise level, wear resistance and resistance to damage by

1
snowplow equipment. In addition, bridge joints should allow easy maintenance, repair, and

replacement. Bridge deck joints fall into two broad categories: open joints and closed joints.

Open joints permit cyclic and long-term movement and support traffic. They may

include armor protection against damage to concrete edges exposed to traffic. The most

common open expansion joints include butt joints either with or without armor facing, sliding

plate joints, and finger joints. The major disadvantage of these joints is that they allow water

and corrosive contamination to pass through them. Therefore, most of the transportation

agencies currently avoid the use of open joints because of this undesirable feature (Purvis,

2003). Draining troughs below the open joints may help ensure that water and debris fall

clear of the supporting structures. However, they produce other set of problems, the most

common ones being the filling up of the trough with the dirt and debris and continuous

maintenance efforts needed to keep them clean and properly functioning.

Sealed or closed expansion joints provide barriers preventing runoff water and

deicing chemicals from passing through the joint on to bearing and substructure elements

below the bridge deck. Water and deicing chemicals have a detrimental impact on overall

structural performance by accelerating degradation of bridge deck, bearing, and substructure

elements (Chen and Duan, 2000). Several types of sealed joint systems have been developed

in an attempt to achieve a joint seal design that would be both effective and durable. Most

widely used among them are hot/cold poured seals, compression seal, strip seal, and modular

joints. These traditional methods of expansion joints have, however, failed to show durability

and longevity and have created much doubt in bridge engineering community as to the

viability of currently in use sealed joint systems.

2
 A number of observations and reports indicate that many deck expansion joint

systems (Both open and closed joints) have performed poorly [for example, see References:

Benson (1986), Chang and Lee (2002), Dahir and Mellott (1985), Frederick (1984), Hamilton

(1985), Voigt and Yrjanson (1992)]. Many have failed to provide water leak proof

conditions. A survey conducted by the Federal Highway Administration (FHWA) (Fincher,

1983) indicated that, in a five year evaluation period, over 60% of the joints were leaking

water and the remaining 40% were experiencing problems that would shorten their service

lives. Where deicing salts are used, leakage of joints promotes the spread of damaging

chloride-laden liquid. Impregnation damage of concrete by chloride is known to be one of

the leading causes of deterioration in reinforced concrete bridges. Under traffic, some of the

joint systems (e.g., finger plate and sliding plate) are objectionably noisy, and the surfaces of

some systems (e.g., compression seal and strip seal) are prone to damage by snowplow

blades (Linfante and Castro, 1977).

Because of the problems associated with leaking joints, the concept of construction of

bridges with continuous superstructure (without intermediate joints) or integral abutment (no

deck joints at abutments) has been introduced in the bridge industry (Burke 1989; Lee 1994).

The use of such constructions is usually justified by the fact that they lead to more

economical bridge design by elimination of costly joints and bearings and they are relatively

easier to repair as the problems of expansion joints are moved from the bridge to its

approaches. However, it is also noted that behavior of such joint- free bridges are still not

completely understood and designs are cumbersome (Thippeswamy et al, 2002)

3
1.2 RESEARCH MOTIVATION

Despite the problems associated with maintenance-prone leaking joints, most existing

bridges have expansion joints, and transportation agencies are concerned with

keeping joints functioning properly and water tight. For new construction, many

transportation agencies are willing to eliminate deck joints from their bridges. This is

achieved by some modifications in the design (e.g., continuous superstructure and integral

abutment) which transfers the problems that take place at joint to another location that might

be off the bridge. Despite the growing tendency of using jointless bridges among

transportation agencies, expansion joints most likely will still continue to remain as an

integral part of many new bridge structures. It is in recognition of this fact and the limited

success achieved with presently used joint sealing systems that this research study was

undertaken as an attempt to develop a new seal for small movement expansion joints that

would be economical, easy to install and maintain, and would accommodate movements over

a reasonable design life while remaining watertight. The new bridge sealant would be

applicable for the use in both new construction as well as rehabilitation work of existing

bridge structures.

1.3 PROJECT OBJECTIVE

The main objective of the research reported in this thesis was to develop an effective

and durable sealing system for small movement bridge expansion joints which can

accommodate total expansion and contraction movements up to 45 mm while preventing the

passage of water, deicing salts, and other road side contaminates through the joint on to the

bridge bearings and substructure components and conduct a series of laboratory tests to

4
evaluate the performance and hence, ascertain its suitability for the real life application. More

specifically the research objective can be summarized in following tasks:

I. To conduct a thorough literature search to identify and evaluate the existing types

of expansion joint systems currently being employed by transportation agencies in

North America

II. To determine factors influencing the successful performance of currently in use

joint sealing systems and provide their relative performance information

III. To develop a silicone based polymeric material as a potential bridge joint sealant in

the laboratory. The newly developed sealant should be economical, easy to install

and repair, and durable

IV. To conduct a wide variety of tests to assess the newly developed sealant’s

mechanical and material properties. Laboratory tests proposed to be conducted

included tension, compression, shear, bonding, salt water immersion, temperature

sensitivity, compression recovery, creep, stress relaxation, cure rate, tack time, and

water tightness test

V. To prepare a final project report that contains research results with

recommendations for phase II (Demonstration and Monitoring Phase) and that

incorporates comments from the NETC Technical Committee.

5
 2. REVIEW OF BRIDGE EXPANSION JOINTS

2.1 BRIDGE JOINTS: GENEREL DESIGN CRITERIA

The specifications for bridge joints are summarized in Section 14 in AASHTO LRFD

Bridge Design Specifications (2004). These specifications are to be followed while selecting

and designing an appropriate expansion bridge joint. State transportation agencies use

AASHTO specifications, with modifications based on their experience and preferences.

Some factors affecting the design and selection of a joint are listed below (Purvis, 2003):

(a) Movement range, (b) Bridge span, (c) Type of bridge, (d) Joint performance and

previous experience, (e) Durability, (f) Maintenance requirements, (g) Bridge alignment, (h)

Joint details at curbs, concrete barriers, or deck edges, (i) Initial cost, (j) Climate conditions

(k) Expected joint life (l) Installation time, (m) Life-cycle costs, (n) Type of bridge supports,

and (o) Service level.

Movement range is the chief concern when designing and specifying a particular

expansion joint. Expansion joints accommodate movements produced by concrete shrinkage

and creep, post-tensioning, thermal variations, dead and live loads, wind and seismic loads,

and structure settlements. Concrete shrinkage, post-tensioning shortening, and thermal

variations are generally taken into account explicitly in design calculations. Because of

uncertainties in predicting, and the increased costs associated with accommodating large

displacements, seismic movements are usually not explicitly included in calculations but

where significant movement is important to the proper function of bridge elements (such as

seismic isolation bearings), the movement due to seismic forces shall be accommodated in

the design of joints. Finally, it may be worthwhile for bridge designers to consider the

performance of expansion joint during lesser intensity seismic events (Gloyd, 1996).

6
 Expansion joints should be designed to accommodate all concrete shrinkage

occurring after their installation. For unrestrained concrete, ultimate shrinkage strain (β) after

installation may be estimated as 0.0002 (Washington State Department of Transportation

(DOT), 2005). Shrinkage shortening of the bridge deck, Δshrink, in mm, is calculated as

Δ shrink = (β )( μ )(Ltrib )(1000mm / m ) --------(1)

Where,

Ltrib = tributary length of structure subject to shrinkage; m

β = ultimate shrinkage strain after expansion joint installation; estimated as 0.0002

in lieu of more refined calculations

μ = factor accounting for restraining effect imposed by structural elements installed

before slab in cast. External restraint to concrete shrinkage is often

provided by the supports of a structural member and by the adjacent structure. The

degree of restraint provided by various supporting systems is incorporated by factor

µ in Eqn. (1). μ = 0.0 for steel girders, 0.5 for precast prestressed concrete girders,

0.8 for concrete box girders and T-beams, 1.0 for flat slabs.

Thermal displacements are calculated using the maximum and minimum anticipated

bridge deck temperatures. These extreme values are functions of the geographic location of

the structure and the bridge type. For example, the temperature range used by Department of

Transportation for the calculation of thermal movement of deck joints is based on a mean

low temperature of -23º C and a mean high temperature of +43º C (Connecticut DOT, 2003).

Thermal movement, Δtemp, in mm, is calculated as (Washington State DOT, 2005)

Δ temp = (α )( Ltrib )(δT )(1000 mm / m ) --------- (2)

7
Where, α = coefficient of thermal expansion; 0.000011 m/m/oC for

concrete and 0.000012 m/m/oC for steel

Ltrib = tributary length of structure in meter subject to thermal variation

δT = temperature variation in oC

Any other predictable movements following joint installation, such as concrete post-

tensioning shortening and creep, should also be included in the design calculations.

2.2 TYPES OF BRIDGE JOINTS AND THEIR ADVANTAGES AND

DISADVANTAGES

Expansion joints fall into three broad categories depending upon the amount of

movement accommodated; (a) Small movement range joints encompass all systems capable

of accommodating total motion ranges of up to about 45 mm, (b) Medium movement range

joints include systems accommodating total motion ranges between about 45 mm and about

130 mm, and (c) Large movement range joints accommodate total motion ranges in excess of

about 130 mm. These delineated ranges are somewhat arbitrary in that some systems can

accommodate movement ranges overlapping these broad categories (Chen and Duan, 2000).

2.2.1 Small Movement Joints

Most common small movement joints include butt joint, sliding plate joint,

compression seal joint, asphaltic plug joint, and poured seal joint. Various features of these

joints are briefly presented below.

8
2.2.1.1 Butt Joint

Butt joint is commonly used for movements less than 25 mm (1 in.). The opening is

provided between two rigid deck slabs with no provision for smooth transition of traffic

between adjacent edges of the deck. They can be built with or without metal armoring angles

as shown in the Fig. 1 (Burke, 1989). A metal angle is typically embedded to protect the top

edge of both sides of the joint from spalling or raveling due to their exposure to continuous

vehicular impacts.

As the butt joint does not provide barrier against water and debris, their use is

precluded in geographical areas where deicing chemicals are used. Another disadvantage

with this type of joint system is that over time, the angles may become dislodged due to the

fatigue of anchor attachments thus creating a traffic hazard.

Figure 1. Butt joint.

2.2.1.2 Sliding Plate Joint

Steel sliding plates, shown in Fig.2 (Washington State DOT, 2005) have been used

extensively in the past for expansion joints in both concrete and timber bridge decks. They

9
are used for movements from 25 to 75 mm (1 to 3 in.) and bridge lengths between joints of

about 110 m (350 ft.). They are structurally simple and reasonable in cost. As seen in the Fig.

2, two overlapping steel plates are attached to the bridge deck, one on each side of the

expansion joint opening. They are generally installed so that the top surfaces of the plates are

flush with the top of the bridge deck. The plates are generally bolted to timber deck panels or

embedded with steel anchorages in to a concrete deck. Standard steel sliding plates do not

generally provide an effective seal against intrusion of water and deicing chemicals into the

joint and on to substructure elements, but it does prevent most debris from passing through

the opening.

Figure 2. Sliding plate joint.

Sliding plate joints are usually limited to horizontal movements and their use is

precluded where differential vertical movements can occur at joints. It is common for plates

to loosen over time. Improperly placed and exposed plates bend, warp and break off from

10
their anchorages due to impact of heavy wheel loads and create a safety hazard (Hill and

Shirole, 1984). Another poor feature of this joint is that the plates need to be adjusted

periodically to reduce noise levels. Due to their unsatisfactory performance, other types of

joints have replaced these joints in most places.

2.2.1.3 Compression Seal Joint

Compression seals, shown in Fig. 3 (Washington State DOT, 2005), are continuous

elastomeric sections, typically with extruded internal web systems, installed within an

expansion joint gap to seal the joint effectively against water and debris infiltration (Chen

and Duan, 2000.) They can accommodate movements from 5 mm to 60 mm (0.25 in. to 2.5

in.). The joint face may or may not be strengthened with armor angles or polymer concrete

header material. Compression seals are held in place by mobilizing friction against adjacent

vertical joint faces. To help fill the voids between the seal and abutting surface, a moisture-

curing polyurethane material with 75 percent solids content is used. Design philosophy

requires that that seal must always be in compression to ensure that it stays in place and

remains watertight. To minimize slippage and maximum compression seal performance, a

joint may be formed narrower than the design width, then sawcut immediately prior to

compression seal installation.

Elastomeric compression seals are versatile, relatively inexpensive and easy to

replace. They have been found suitable for bridge application by many transportation

departments, and are given very high ratings by these departments; however, other

departments have abandoned their use entirely (Purvis, 2003). Agencies such as New York

and Illinois, report that these are effective seals that require minimal maintenance with a

11
reasonable life span. Others such as Louisiana and Colorado, report unreliable performance.

These seals are very much susceptible to damages from snowplows, debris, and traffic. For

wider joints, it is difficult to ensure the adherence of the seal to the sides of the joint. Over

time, the compression seal gradually looses its capability to retain its initial compression

recovery due to loss of resilience and becomes brittle. This leads to leakage of water in to the

bridge substructure through joint-seal interface causing accelerated deterioration of the

substructure, bearings, and superstructure beams under the joint (Fig. 4; Ohio DOT, 2003).

Figure 3. Compression seal.

12
 ¾Seal become brittle with age
¾Adhesive failure

Figure 4. Compression seal leakage.

2.2.1.4 Asphaltic Plug Joints

Asphaltic plug joint (APJ) is the most popular joint sealing system in New England

transportation departments. The system is becoming popular with agencies in other parts of

the country as well. Movement capacity of such joint is less than 50 mm (2 in.). Typical

components of the joint are shown in Figure 5 (Washington State DOT, 2005).

It consists of a blockout 570 mm wide by 60 mm (minimum) deep (20 in. x 2in.). A

closed cell, heat resistant backer rod is placed in the joint below the blockout. The joint is

filled above the backer rod with polymer modified asphalt binder material (PMA), and a

traffic bearing steel plate 230 mm (8 in.) wide by 6 mm (0.25 in.) thick is centered over the

joint, bridging the opening, for its entire length. The blockout is filled with clean, dry, open-

graded aggregates coated with the asphalt binder. After placement material is consolidated

with a vibrating plate compactor and binder material is poured over the top of the compacted

material to fill the voids.

13
 Figure 5. Asphaltic Plug Joint.

The primary advantage of this system is ease of installation and repair. In addition, it

provides smooth and seamless roadway surface to the traffic. There is no debris collection

atop the joint and they provide watertight and snowplow proof expansion joints. The major

disadvantage of the system is that it is not effective method of sealing vertical joints or

skewed joints. At locations with very high temperature, PMA material softens and creeps,

which leads to wheel rutting and migration of binder from blockout [Fig. 6 (Ohio State DOT,

2003)]. Similarly, cracking of PMA occurs in very cold weather. Daily and seasonal

temperature variations are stress inducing events for the APJ material. If the relaxation of the

APJ material is not sufficient, the significant stresses are developed in the joint causing crack

at joint-to-pavement interface (Bramel et al, 2000). Figure 7 (Ohio State DOT, 2003) shows a

section coming out of the failed plug joint.

14
 ¾Improper installation
¾Unsound concrete below
¾High binder content

Figure 6. Material flowing on to adjacent pavements.

¾Improper compaction
¾Low binder

Figure 7. Section of failed joint.

15
 2.2.1.5 Poured Sealed Joints

Durable low-modulus sealants, poured cold to provide watertight expansion joint

seals as shown in Fig. 8 (Washington State DOT, 2005), have been used in new construction

and in rehabilitation projects. Traditionally, such sealants are used on shorter spans where

joint movement is up to 6 mm (0.25 in.). However, many new sealants have been developed

and claimed by the manufactures for larger movements. The sealants with movement rating

+100/-50% are not uncommon today.

Figure 8. Polymer poured-in-place seal.

The system typically includes an elastomeric header with pourable silicone sealer and

polyethylene backer rod as joint filler. The backer rod is squeezed into the joint to prevent

the spilling of the sealant through the joint opening and to form the required sealant shape.

The silicone is a self-leveling, rapid curing, one-part or two-part polymeric material. Most

silicone sealants possess good elastic performance over a wide range of temperatures while

demonstrating high levels of resistance to UV and ozone degradation. Rapid curing sealants

16
are ideal candidates for rehabilitation in situations where significant traffic disruption from

extended traffic lane closure is unacceptable. The major problem associated with this type of

system is debonding of the seal material from the joint face [Fig. 9 (Courtesy New

Hampshire State DOT)]. Other problems including splitting and damage from non-

compressible debris have also been reported.

Figure 9. Debonding of silicone sealant.

2.2.2 Medium Movement Joints

Strip seal and finger joints are most popular medium movement joints.

2.2.2.1 Strip Seal

A strip seal consists of “V” shape preformed neoprene gland mechanically locked

into a metal facing on both sides of the joint. Movement is accommodated by unfolding of

17
the elastomeric gland. Movement capacity of strip seal is up to 100 mm (4 in.). A typical

detail is shown in Fig. 10 (Washington State DOT, 2005).

Figure 10. Strip seal.

Properly installed strip seal systems are watertight and have demonstrated relatively

better performance. Another benefit is that damaged or worn glands can be replaced with

minimal traffic disruptions. Many state agencies such as Minnesota, New Jersey, and

Pennsylvania have reported better and more durable performance of strip seals with respect

to other joint seals. The most prominent disadvantage of this system is that the elastomeric

gland exhibits a proclivity for accumulating debris. Incompressible materials lodged in

membrane crevices resist joint movement and results in premature gland failure [Fig. 11

(Ohio State DOT, 2003)]. Additionally, faulty installations or unclean locking devices cause

gland pullout from metallic rail edges [Fig. 12 (Ohio State DOT, 2003)].

18
Figure 11. Tearing of seal due to incompressible debris.

Figure 12. Gland missing.

19
 2.2.2.2 Finger Plate Joints

These joints are generally fabricated from steel plate and are installed in cantilever or

prop cantilever configurations [Fig. 13 (Burke, 1989)]. The joints accommodate rotational

movement in addition to differential vertical deflection across the joint. Generally, drainage

trough is below the joint to intercept deck water and debris and carry it away from

substructure members.

Agencies such as Arkansas and Maine have given high ratings for finger joints.

However, open finger joints without drainage trough were no longer specified in Maine.

Where narrow bicycle tires are anticipated, floor plates should be used in the shoulder area.

On evaluation of bridge joints for Pennsylvania, Dahir and Mellott (1985) commented that

cantilever fingers can be bent or broken under continuous pounding of heavy traffic. Also,

due to rapid accumulation of cycles of fatigue loadings, welding details are critical for these

systems.

Figure 13. Finger Joint. (a) Plan view; (b) Section A-A.

20
2.2.3 Large Movement Joints

Large movement joints are used for accommodation movements greater than 100 mm

(4 in.). Most common among them are plank seal and modular joint.

2.2.3.1 Plank Seal

The plank seal consists of monolithically molded elastomeric panels reinforced with

steel plates as shown in Fig. 14 (Burke, 1989).

Joint Width

Figure 14. Plank seal

The plank seal is used for movement ranges from 50 to 330 mm (2 to 13 in.). These

joints have lost favor among the agencies that use snowplows. Snowplows cut into the

material and occasionally destroy complete sections. Cost is an important consideration with

the system as complete replacement of the joint is required when damaged. Anchor failure is

another problem. Bolts and nuts connecting panel to bridge decks are prone to loosening and

breaking under high-speed traffic. Figure 15 (Purvis, 2003) shows an example of plank seal

failure.

21
 Figure 15. Damaged plank seal.

2.2.3.2 Modular Joints

Modular bridge expansion joints, shown in Fig. 16 (Connor, 1999), are complex,

expensive structural systems designed to provide watertight wheel load transfer across wide

expansion joint openings. The joint components are sized according to the magnitude of

movement. Typical movements associated are 150 mm (6 in.) to 600 mm (24 in.). The

system comprises of a series of center beams supported atop support bars. The center beams

are oriented parallel to joint axle while support bars span parallel to movement direction.

Fatigue cracking of welds, damage to neoprene sealer material, and damage from

snowplows are major problems with this type of system. An example is shown in Fig. 17

(Purvis, 2003). Most agencies are reluctant to use modular joint on account of their high

initial costs and high maintenance cost.

22
Figure 16. Modular joint.

Figure 17. Damaged modular joint.

23
2.3 BRIDGE EXPANSION JOINTS IN NEW ENGLAND STATES

Transportation agencies in New England states were contacted and information on types

of expansion joint systems they have in their state bridges, along with relative performance

of each type of system, was collected. Based on their responses, the various types of joint

systems currently in use in New England states and their range of applicability and

performance are summarized in Table 1 below.

Table 1. Various Bridge Deck Joints in Practice in New England

State Types of Joints Anticipated Movement Comments
Employed Range (MR) or
Deck Span Length (L)
Connecticut a. Asphaltic Plug Joint MR < 40 mm 95 % of all joints
b. Silicone Sealant MR: 40-80 mm Elastomeric header
c. Neoprene Strip Seal MR: 80-100 mm Elastomeric header
d. Modular and Finger Plate MR > 100 mm -
Maine a. Compression Seal - Most preferred
b. Silicone -Pour-in-Place Small MR Rehabilitation project
c. Gland Seal MR > 100mm -
d. Evazote Seal - Limited success
e. Asphaltic Plug Joint MR < 50mm No success, Failure in
short period
Massach- a. Saw Cut Seal L < 15 m -
usetts b. Asphaltic Plug Joint L > 20m, <35m Skew < 25º
c. Strip Seal L > 35 m Armored
d. Finger Joint Large spans Neoprene trough

New a. Silicone based Sealant Small MR Reasonable success

Hampshire b. Roadway Crack Sealer For short spans and Hot applied, petroleum
on fixed ends based
c. Asphaltic Plug Joint L: 80’-140’ Good results, skew <25º
d. Finger Joint L: 140’-180’ -
Rhode a. Compression Seal - Poor perfromance, No
Island more in use
b. Strip Seal Large MR Poor performance,
Leakage
c. Asphaltic Plug Joint Short Spans (L<100’) Most preferred
d. Open Joints, Sliding - Exist in old construction
Plate Joint
Vermont a. Asphaltic Plug Joint MR: 50-75mm; Most preferred
Short Spans (L<90’)
b. Vermont Joint MR < 75mm (L>90’) -
c. Finger Plate Joint MR > 75mm -
d. Modular Joints Very Large MR. Rarely used

24
 From the data provided in Table 1, asphaltic plug joint seems to be the predominantly

used expansion joint system in New England region. They are normally used to

accommodate movement ranges of up to about 50 mm with a skew angle less than 25º.

2.4 EVALUATION OF PERFORMANCE OF EXPANSION JOINTS

2.4.1 Common Joint Defects

Since installation, joints are attacked not only by natural elements such as water, dust,

dirt, ultraviolet rays, and ozone, but also elements introduced by humans such as deicing

chemicals, snowplows, or traffic impacts. Each possible defect is a contributing factor to

deterioration. The common joint defects are as follows (Guzaltan, 1993):

1. Loose, torn, split, cracked, damaged, or hardened seal;

2. Accumulation of debris and incompressible materials in the seals, drainage troughs,

downspouts, and silting basins;

3. Loose, rusted, cracked, missing, or damaged steel plates, shapes, anchorage, bolts,

nuts and other metal components;

4. Cracked and spalled concrete and rusted or exposed reinforcement steel or

structural steel in deck joint substrate;

5. Water leakage and its impact on the underside of deck;

6. Impact of noise during passage of vehicles over joint;

7. Restriction of freedom of joint movement;

8. Impact of rotation, tilting, or settlement; and

9. Incorrect joint opening or improper joint clearance and alignment.

25
2.4.2 Factors Influencing Joint Performance

The performance in service of many joints has been variable, and the reasons for this

are not readily apparent. A survey conducted by Transportation and Road Research

Laboratory identified that a wide range of factors influence the performance of joints (Price,

1984). They include:

1. Structural movements at joint,

2. Traffic loading,

3. Joint Design,

4. Materials used,

5. Detritus, foreign matter, and corrosion,

6. Bond and anchorage,

7. Condition of substrate,

8. Weather and temperature during installation and service,

9. Detritus, debris, and corrosion,

10. Site preparation and workmanship,

11. Performance of bearings

These factors differ between and within joint types and frequently it is a complex

combination of factors which affect serviceability and not necessarily the same combination

or sequence for all the joints. Often the action of one factor instigates another. For example,

traffic density and axle loading may influence the performance of bearings, and site

preparation and workmanship could influence the quality of bond and anchorage, all of

which influence the performance of joints.

26
2.4.3 Bridge Deck Joint Evaluation

A significant amount of literature is available on performance studies of bridge deck

expansion joints and sealant systems [For example, see Biel and Lee (1997), Chang and Lee

(2002), Eacker and Bennett (2000), NCHRP (1979)]. These studies usually involve either

questionnaire survey to determine the level of success that transportation departments have

experienced in the use of various types of bridge deck joints or the field inspection of joints

at selected test sites. The most frequently encountered problems and their causes and the

merits of joint types are identified from the survey which is used to rank the performance of

these joints.

Figure 18 shows a sample of completed evaluation form used in bridge deck joint

evaluation survey conducted by Transportation Research Board (Burke, 1989). Joint types

included in survey were open joints, finger joints, slider joints, compression seals, strip seals,

sheet seals, plank seals, cellular seal, and modular compression seals. Figure 19 shows the

summary of the evaluations contained on the forms received from 31 transportation

departments who participated in the survey. The summary shows the relative success or

failure that has been experienced by transportation departments in the selection and

application of various types of bridge deck joints.

Field performance of three types of joint sealants, namely silicone sealants, PVC-coal

tar sealants, and hot-pour sealants were compared in a study performed for Utah Department

of Transportation (Loza et al, 1987; Voigt and Yrjanson, 1992). The sealants were installed

and their performances were evaluated after one, two, three, and eight years. After two years

silicone materials were observed to be performing well (4% failure) and the sealant continues

27
to perform well after eight years. The PVC-coal tar was having adhesion problems, and the

hot–pour rubberized asphalt materials performing poorly having up to 80 % failure.

Figure 18. Sample evaluation form.

28
Figure 19. Summary of transportation departments’ evaluation of bridge deck joints.

29
 A study conducted for Indiana Department of Transportation (Chang and Lee, 2002)

investigated the performance of compression seal, strip seal, integral abutment, poured

silicone, and polymer modified asphalt (PMA) joints. Besides the usual questionnaire survey,

the study used the logistic regression approach to analyze the Indiana Department of

Transportation (INDOT) roadway management data to evaluate the joint performance. The

final joint performance ranking based on parameters such as age, traffic volume, and

settlement, are shown in Table 2. Higher ranking indicates the slower deterioration rate under

the influence of the factor. Due to the small number of data, poured silicone was not included

in the analysis. Similarly, PMA joint could not be compared with other types of joints

because no common factor was selected for the PMA joint.

The study indicated that the strip seal joint performed the best, compression seal

second and integral abutment joints third, under the conditions of age and settlement. The

strip seal joint performed better than the compression seal joint under the conditions of traffic

loading.

Table 2. Joint Performance Ranking Based on Deterioration Rate

Variable

Ranking Age Traffic loading Settlement

1 Strip seal Strip seal Strip seal
2 Compression seal Compression seal Compression seal
3 Integral abutment - Integral abutment

The evaluation study performed by Virginia Department of Transportation (VTDOT)

that focused on joint sealing systems showed mixed results with regard to the performance of

the generic systems evaluated (French and McKeel, 2003). The research procedure employed

in this evaluation was a series of case studies of individual trials of joint sealing systems used

30
by VDOT during the past several years. It did not include butt, sliding plate, and finger

joints. Larger modular joint systems were also excluded. Each system was found to perform

successfully at one occasion, while showing a failed installation at other. The study

concluded that adherence to recommended installation procedures is essential to attaining

satisfactory service from a joint sealing system.

A study conducted by NCHRP (Purvis, 2003) showed that most of the closed joints

start leaking water within a period of 5 years from the joint installation. This leakage was

attributed to the following factors:

(a) Improper Installation, (b) Poor maintenance, (c) Puncture/damage by debris, (d) Defective

product, (e) Damage by traffic, (f) Damage by snowplow, and (h) Temperature extremes.

Debris collection was found to be another problem with deck joints. Almost all of the

agencies who took part in the study reported that their deck joints commonly collect debris,

with 80 % noting that the debris had adverse effects on joint performance. The study listed

following other performance problems for bridge deck expansion joints:

• Compression failure when joint closes,

• Snowplow damage,

• Substructure movement, and

• Poorly formed headers.

31
 3. DEVELOPMENT OF NEW SEALANT MATERIAL

3.1 SELECTION OF JOINT SEALANT MATERIAL

After reviewing the existing literature on expansion joint sealants, silicone based

polymeric material was selected as the potential new joint sealant. Silicone sealants for

concrete are well-established as the Cadillac of sealant systems. As such, they have long life,

flexibility under all temperature conditions, outstanding moisture resistance and reasonable

UV resistance, which can be improved with additives, and outstanding adhesion to a wide

variety of surfaces.

Silicone chemistry is extremely versatile, with both one- and two-part room-

temperature-curing formulations available, as well as one or two part thermally curing

varieties. However, there are some important drawbacks of silicone sealants. Cost is high,

relative to rubber-modified asphalt or even polyurethanes. In addition, the strength of the

elastomer is low, so resistance to damage by the penetration of pebbles is limited. One

possible route to improving on these two properties is to switch to silicone foams [e.g.,

Stadelmann (2001), Vincent (1972)]. Foams can be made using two-part systems. The

normal method of foaming employs a two-part formulation with one ingredient containing a

hydrosilane and the other an alcohol. When mixed, the reaction proceeds as follows:

-(CH3)2Si-H + ROH Æ -(CH3)2Si-OR + H2

with the hydrogen gas forming the bubbles resulting in a closed-cell foam (Dubiel et al.

1983). The alkoxy silane group is then available for reaction with water on the surface of the

concrete to form silanol, which then condenses to form a silicone elastomer coating on the

concrete. The hydrosilane can also react with silanol groups to form hydrogen and extend

32
the length of the chain. As this sealant material expansion can be high, cracks can be filled

with less material, which will lower cost and partially compensate for the generally higher

cost of the hydrosilane-functionalized prepolymers. The foam structure can often localize

the damage created by penetration, and its low stiffness will reduce the transfer of shear

stress to the fragile concrete-polymer interface. Spot repairs should be quick and easy.

Following requirements were kept in mind while developing the new silicone foam

sealant material in the laboratory:

• Movement capability: sealant should be able to accommodate total contraction and

expansion movements up to 1.5 inches, combined with shear stress to simulate

skewed bridges

• Elasticity: sealant should be able to return to its original size after released from

elongation or compression

• Modulus: low resistance to load at low temperatures. The lower resistance at lower

temperatures is achieved through more flexible material formulation and keeping

lower glass transition temperature

• Adhesion: sealant should be able to adhere well to a variety of joint substrates, such

as concrete and steel

• Cohesion: material should be strong enough to resist tearing from tensile stress

• Compatibility: chemical reaction of a sealant to material that it contacts

• Weatherability: able to resist deterioration when exposed to natural elements such

as ultraviolet sun rays and ozone

• Applicability: able to be applied at ambient temperature

33
 • Reparability: should facilitate easy removal of damaged section and replacement

with a new section able to bond with the existing system

3.2 LABORATORY DEVELOPMENT OF SILICONE FOAM SEALANT

Silicone foam sealant was prepared in the laboratory by adding Baysilone U430

crosslinker (GE Bayer Silicones, 2003), deionized water, and platinum catalyst (Gelest Inc.,

2003) to the mixture containing equal amount by weight of Wabo® Silicone seal grey and

white components (Malla et al. 2005a, 2005b). The percentage weights of crosslinker, water,

and catalyst used for the silicone foam sealant preparation were 2.3, 1.53, and 0.39 %

respectively with respect to the total weight of Wabo® Silicone seal grey and white mixture.

Figure 20 presents a schematic representation of the chemical reaction for the foam

formulation. Si-H in Wabo® Silicone seal and cross linker reacts with water in presence of

platinum catalyst to form silanol and hydrogen gas (H2). Si-OH of silanol can react with

another Si-OH or with the crosslinker to further polymerize. The released hydrogen –leads to

foaming of the sealant. The foaming of the sealant results in significant rise in the sealant

volume. Preparation of foam test specimens showed 70% increase in the sealant volume after

the chemical reaction. This is highly desirable feature as it will lead to the considerable

saving in the sealant material during joint sealing operation and will force the sealant into

areas with poor access. The volume rise was determined by preparing two types of sealant

specimens: solid sealant specimen containing Wabo® Silicone seal grey and white

components only, and foam sealant specimen containing crosslinker, water, and platinum

catalyst in addition to the solid sealant grey and white components. The volume occupied by

these two sealant types were measured on complete of cure reaction and rise was calculated.

34
 Pt catalyst

U430
Crosslinker

Figure 20. Schematic representation of foaming reaction.

As the reaction starts upon the introduction of the Pt to the vial so it is critical to make

laboratory samples quickly. Rapid stirring allows for sufficient mixing in under 30 s but the

sample must be injected into the joint as quickly as possible because the foam will not rise as

high if too much hydrogen is “wasted” during mixing and injecting. The foam sealant has

an average experimental density of 0.064 grams/cm3 and expands to nearly 1.7 times their

original volume. Preparation of foam involves a short laboratory working time; however this

issue will not be a problem in the field with proper installation equipment.

It should be noted that the newly developed sealant material, termed as the foam

sealant hereinafter, forms a foam structure on curing (Fig. 21a) unlike the source material

Wabo® Silicone seal, termed as the solid sealant hereinafter, which forms a solid rubber (Fig.

20b). It should also be noted that both foam and solid are “elastomers”, a term often used for

rubber and polymers that have properties similar to those of natural rubber. Elastomers are

polymers with a particular molecular structure. Specifically they are linear, amorphous, high

molecular weight polymers in which the chains of molecular structure are flexible and cross-

linked (Hearle, 1982). The flexibility and cross linking of the polymer chains allow

35
elastomers to behave elastically, as the name implies, such that they are capable of

recovering from large deformations.

a.

Figure 21. Two silicone seal materials

(a) Silicone foam sealant, (b) solid sealant.

36
 4. LABORATORY EVALUATION TEST METHODOLOGY

To determine the characteristics and evaluate the performance of the newly developed

foam sealant, several important laboratory tests were proposed to be conducted. They

included: tension, compression, shear, bonding, salt water immersion, temperature

sensitivity, compression recovery, creep, stress relaxation, cure rate, tack time, and water

tightness test. Some of the abovementioned tests were repeated on weathered sealant after its

exposure to outdoor condition for 6 months. Similar tests were also proposed to be performed

on a currently available commercial solid sealant (Wabo® Silicone Seal, Watson Bowman

Acme Corp. 2003) intended to be used as a control during laboratory validation testing of the

foam sealant. Preparation of test specimens and laboratory test methods are presented in this

chapter.

4.1 TENSION TEST

4.1.1 Loading Test

Figure 21 shows the typical tension test specimen configuration. The test specimen

consisted of two concrete blocks with a gap in between them that was filled with sealant.

Dimensions of each concrete block were: depth, D = 12.7 mm (0.5 in.); width, W = 50.8 mm

(2 in.); length, L = 76.2 mm (3 in.). The dimensions of the sealant gap were: depth, D = 12.7

mm (0.5 in.); width, W = 50.8 mm (2 in.); length, ι = 12.7 mm (0.5 in.). Three such test

specimens were prepared for each sealant type (foam sealant and solid sealant) to be tested.

The foam sealant was prepared by the procedure described in the preceding section and

poured into the gap formed between the concrete blocks. The concrete substrates were wiped

37
with a clean, dry, lint-free cloth prior to pouring of the sealant; however, the joint faces were

not primed. Solid sealant specimens were prepared by hand mixing equal parts by weight of

Wabo® Silicone seal grey and white components and then pouring the mixture into the gap

between the blocks. Table 3 shows the different material components and the amounts used

to prepare a tension test specimen. Whenever necessary, the sealant surface was tooled to

ensure complete filling and contact with substrate surfaces. The test specimens were cured

for 21 days at 23 ± 2°C (room temperature). The relative humidity, however, was not

monitored and maintained. To simulate actual field conditions, no elevated-temperature post

cure was used.

Figure 22. Tension test specimen.

38
 Table 3. Sealant Components for Tension Test Specimen
Components Foam sealant Solid sealant
(gram) (gram)
Wabo grey 3.50 6.50
Wabo white 3.50 6.50
Baysilone
0.1615 None
U430 crosslinker
Water 0.107 None
Pt-divinyltetramethyl-
disiloxane complex 0.027 None
(catalyst)

An Instron tensile tester (Model 1011) (Fig. 23) was used for testing the specimens

prepared above. The machine was capable of producing uniform rates of grip separation

varying from 1 mm/min to 500 mm/min. The load cell was connected to a computer for

recording the applied force. The crosshead movement was used to calculate the elongation.

The test specimen was placed in the grips of the testing machine, using care to adjust the

specimen symmetrically to distribute tension uniformly over the sealant cross section. Due to

the low aspect ratio of the specimens, the state of stress is not simple extensional, but because

of the high compliance of the material, this was judged to have minimal effect on the results.

The specimens were then pulled at a constant crosshead velocity of 10 mm/min (nominal

strain rate of 0.013 s-1) until the failure occurred (Fig. 24). Failure here refers to either tearing

apart of the sealant material (Cohesive failure) or detachment of the seal from the concrete

block substrate (Adhesive failure). The nominal stress and strain developed in the sealant

material were calculated using the following mathematical expressions (Eq.3 and Eq. 4).

Ρ Δ
σ = ---------- (3); ε = × 100 % --------- (4)
Α ι

39
Where, σ = Average normal stress (nominal); P = Applied tensile force; A = Unstrained

sealant cross sectional area; ε = Normal strain (nominal); Δ = Sealant extension; and ι =

Unstrained sealant length

Figure 23. Instron testing machine (Model 1011) with tension specimen.

4.1.2 Loading and Unloading Test

The loading and unloading behavior of sealants in simple tension was also studied.

The tension test specimens were loaded in the Instron machine up to the elongation of 300 %

strain at the rate of 10 mm/min, and then unloaded at the same rate until they reached their

original length (Zero strain). The procedure was repeated for five consecutive cycles of

loading and unloading.

40
 Sealant

Concrete Blocks

Sealant

Figure 24. Tension test of sealant using Instron machine.

4.2 COMPRESSION TEST

4.2.1 Loading Test

A sealant block 76.2 mm (3 in.) by 25.4 mm (1 in.) by 12.7 mm (0.5 in.) was cast and

cured at standard laboratory conditions (23 ± 2° C) for 7 days. After curing, test coupons

were cut from the sealant to produce 3 uniform test samples each of size 25.4 mm (1 in.) by

25.4 mm (1 in.) by 12.7 mm (0.5 in.). Three such test samples were prepared for each type of

sealants (foam sealant and solid sealant). The upper and the lower grips of the Instron testing

machine were replaced with a cylindrical loading head and a circular bearing plate,

respectively. Test specimen was carefully placed on the bearing plate so as to align its axis

with the center of thrust of the cylindrical loading head. All specimens were uniaxially

41
compressed at a uniform strain rate of 10 mm /min to the half of their thickness (50% strain)

and the modulus at this strain was obtained. A schematic diagram of the test set up is

presented in Fig. 25.

Cylindrical loading
head

Sealant block

Circular bearing
plate

Figure 25. Compression of sealant.

4.2.2 Loading and Unloading Test

As in the tension test, sealant behavior under compression loading and unloading

cycles was also evaluated. The specimens were compressed using the Instron machine to half

of their original thickness (50 % strain) at the rate of 10 mm/min, after which the load was

released by reversing the motion of the loading head at the same rate till the strains in the

specimens were 0%. The procedure was repeated for 5 consecutive cycles of loading and

unloading during which force values were constantly measured.

4.3 SHEAR TEST

Figure 26 shows shear test specimen in double lap joint configuration (sandwich

arrangement) used in this study. As shown, the middle moving block shears two sealant slabs

42
between two fixed blocks. The sealant layer dimensions on each joint were as follows:

thickness, t = 6.3 mm (0.25 in.); height, h = 25.4 mm (1 in.); width, b = 50.8 mm (2 in.)

(Direction out of paper). This gives sealant-substrate contact area on each side equal to 1290

mm2 (2 in.2). The inset in Fig. 26 shows the free-body diagram of the deformed sealant under

the applied load P. Three test specimens for each sealant type were prepared and cured for 14

days at 23 ± 2ºC. Cement mortar substrates were used in contact with the sealant surface.

P
Sealant

h
t Δ θ
P/2
P/2

Cement Mortar Blocks

Figure 26. Shear test specimen.

Instron testing equipment described in section 5.1 was used for the shear test. As

shown in Fig. 27, the upper lower grip was removed and the shear sandwich was placed

symmetrically below the movable loading head. The center block was pushed downward 25

mm/min (strain rate of 0.066 s-1) until the sample failed. The shear stress (τ), shear strain (γ),

and elastic shear modulus (G) of the sealant were determined using the following

relationships (Eq. 5, Eq. 6, and Eq. 7).

43
 Ρ Δ τ
τ= ------- (5); γ = (tan θ ) = ------- (6); G = ------(7)
2Α t γ

Where, P = Applied Load; A = Sealant cross sectional area in contact with each substrate;

Δ = Vertical deformation; t = Sealant thickness; θ = Shear angle

Loading Head
Concrete Blocks
Foam Sealant

Figure 27. Shear test of foam sealant using Instron machine.

4.4 SALT WATER IMMERSION TEST

Sealants were exposed to saturated solution of sodium chloride (common salt) in

water at an elevated temperature of 45 ºC for 2 weeks to determine the influence on their

adhesion and moisture resistance properties. The elevated temperature of 45 ºC was

maintained to facilitate the accelerated weathering and chemical effects. The primary effects

due to moisture may include the absorption of water by the sealant and the diffusion of liquid

44
water to the sealant substrate interface. The absorption of water may cause either softening or

enhanced cure of the sealant, whereas the diffusion may result in the impairment of the

adhesive bond of the sealant to the joint surface (Beech, 1988) and damage during freeze-

thaw cycles. The effects due to water absorption can be evaluated by means of tension testing

in which the modulus and extensibility of sealants after immersion are measured as indices of

performance (Beech and Mansfield, 1990). Effects on bonding due to immersion can be

evaluated from the failure modes observed during the tests.

Three specimens of each type of sealants, foam and solid sealant were prepared using

concrete substrates as described in tension test. After a week’s cure at room temperature, all

specimens were immersed in salt water at an elevated temperature of 45°C. After two weeks

immersion, the specimens were taken out, allowed to dry in air for 3 to 4 hours, and then

pulled to the rupture using the steady cross head velocity of 10 mm/min.

4.5 BOND (OVEN AGED) TEST

Adhesion to the substrate is a critical property of joint sealant. The test procedure as

suggested by ASTM D 5893-96 specification (ASTM, 1997) was employed to evaluate the

bond between the sealant and the concrete substrate. Oven aged samples were subjected to

alternate cycles of cold extension and self-recompression and then inspected for any cracks

and separations within the sealant or at the sealant-substrate interface. Specimens for this test

were prepared in the same way as the tension test specimens. Three test samples of each type

of sealants (foam sealant and solid sealant) were prepared and cured at standard laboratory

conditions (23 ± 2° C) for 7 days, which was followed by oven aging of the samples at 70 οC

for a week. The oven-aged samples were then put in a cold chamber maintained at -29οC for

45
4 hours before mounting them to the grips of Instron testing machine. The maximum and

minimum temperature values have been chosen as the test temperatures as they are close to

the expected upper and lower service temperature of most bridge joint systems in the U.S.A.

The samples were extended at uniform rate of 6 mm/min to the length twice their original

length (100 % strain), and then removed from the testing equipment and allowed to regain

their original length at room temperature for 2 hours. The self-recompressed samples were

again put in a cold chamber at -29οC and whole procedure was repeated for 5 cycles of

extension and self-recompression. After the 5th extension, the test specimens were removed

from the extension machine and immediately examined for obvious separation within the

sealant and between the sealant and the blocks.

In addition to the visual evaluation of bonding at the sealant joint interface, the rate of

loss of modulus with deformation cycles was also analyzed. To compare more realistically

the modulus loss rates of two sealants with deformation cycles, a classical expression

proposed by Chasset and Thirion (Thirion and Chasset, 1967) was used. The original

Chasset-Thirion (C-T) equation was applied to stress relaxation of crosslinked elastomers

and was explained in terms of the disentanglement of network chains that were attached at

only one end to the network. The simplified expression (Eq. 8) used to calculate modulus

after any number of cycles was:

E = E∞ [1 + (a/n) m ] --------- (8)

Where E∞ is the modulus after infinite number of cycles, n, and a represents the number of

cycles that will result in a modulus of 2 E∞.

46
4.6 STRESS RELAXATION TEST

When an elastomer is subjected to constant strain under tension, compression or

shear, the stored energy in the material decreases over time. This stress relaxation

phenomenon occurs in elastomeric material due to two distinct effects, the first physical (due

to viscoelasticity) and the second chemical (due to aging of the rubber such as by bond

exchange or chain scission). At short times and low temperatures, physical effects are

dominant whereas chemical effects are more apparent at longer times and higher

temperatures (Smith, 1993).

Joint sealants operate both in tension and compression and hence stress relaxation

measurements in tension as well as compression were used to measure sealing efficiency. To

measure decay of sealant stress with time in tension, the tension specimen was pulled in an

Instron machine to 100% strain and maintained at this stretched condition for a 24-h period.

To determine the sealant relaxation behavior in compression, sealant samples were

maintained at 50% compression strain using the Instron machine for a 24-h period. In both

cases, the decrease in stress with time was monitored and recorded during test period. The

stress measurements can be normalized to the initial stress measurement and expressed as

percentage using the expression (Eq. 9):

Rt = (σ0 – σt)/σ0 x 100 --------- (9)

Where Rt is the stress relaxation after time t, σ0 is initial stress, and σt is stress at time t.

The relaxation phenomena in the sealants were found to follow a Kohlrausch-

Williams-Watts stretched exponential function (KWW equation). The original KWW

equation was used to describe the dispersive relaxation phenomena observed by Kohlrausch

47
in a study of the loss of charge in Leyden jars (Kohlrausch, 1863). Williams and Watts

(1970) described dielectric relaxation in polymers as being a stretched-exponential function.

The stretched exponential form of the relaxation modulus E(t) used to reduce experimental

data (Eq. 10) was:

E(t) = (E0 - E∞) exp[-(t/τ)β] + E∞ ----------(10)

Here 0 < β < 1, t is time, and E0 and E∞ are constants. The parameters representing relaxation

time (τ) and stretched exponential constant (β) depend on the material and can be a function

of external variable such as temperature (Klafter and Shlesinger, 1986). These parameters

were calculated according to the Eq. 10 using SigmaPlot 9.0 graphing software.

4.7 CREEP TEST

Creep is the increase in deformation after a specified time interval under a constant

load and expressed as a percentage of the test piece deformation at the commencement of

that time interval (Smith, 1993). Hence,

Creep % (εcreep) = (Dt – D1)/ D1 x 100 -------- (11)

Where, Dt = Deformation of the test piece after t minutes and D1 = Instantaneous deformation

of the test piece.

As in stress relaxation, creep may be due to physical (at low temperature and short

times) and chemical (at high temperature or long times) effects. Tensile test specimens were

subjected to constant load and change in deformation with time were noted for 24-h period.

Magnitudes of constant load for foam and solid were 11N and 30N respectively. These

48
values were chosen so as to cause sealants to undergo instantaneous elongation of

approximately 100% upon the application of load.

4. 8 COMPRESSION RECOVERY TEST

During hot weather, an increase in temperature causes bridge deck slabs to expand.

As a result, the joint starts to close and the sealant gets compressed. Sealants in bridge joints

are typically expected to accommodate compressive movements of up to 50% of joint width.

During prolonged compressed state, the stresses within the sealant may get relaxed which can

make the sealant loose its ability to return to its original size and shape later when joint

widens due to temperature decrease. This may cause the sealant to fail cohesively or/and

adhesively due to higher tensile stresses developed as the joint opens. The scope of this test is

to measure the sealant’s ability to recover its original state after being subjected to

compression under high temperature for a 24-h period. The test method suggested by ISO

815 (ISO, 1991) was used to determine the compression set at elevated temperature with only

exception being the geometry of the test specimen. The test specimen consisted of a block of

the sealant with 25.4 mm by 25.4 mm square cross section and 12.7 mm thickness which was

placed between 2 parallel 50.4 mm by 50.4 mm faces of similar concrete blocks. The

specimen was then compressed to 50% of the original thickness and placed in an oven

holding it in compression at an elevated temperature of 45º C for 24 h. At the end of test

period, the test piece was released from compression and its thickness was measured at

different time intervals until the measured thickness reached more or less a constant value.

The compression set (εset) is the difference between the original thickness of the test piece

49
 and that after recovery, as shown in Fig. 28, expressed as a percentage of initially applied

compression (Eq. 12).

Compression set% (εset) = (t0 – tr)/ (t0 – ts) x100 -------- (12)

Where, t0 = original thickness of the test piece, tr = thickness of test piece after recovery,

and ts = thickness of test piece after initial compression

Degree of
Permanent Set (εset)

Specimen before compression Compressed specimen Specimen after recovery

Figure 28. Diagram illustrating compression set.

4.9 TEMPERATURE SENSITIVITY TEST

4.9.1 High and Low Temperature Effects

Silicone sealants exhibit excellent thermal stability (Stoegbauer and Wolf, 1990).

Resistances of up to 200 οC and, for special formulations, even 250 οC are quoted in the

relevant literature. Cold-period loading of a seal is generally recognized as the most critical

loading period because of possible sealant hardening at low temperatures. Below the glass

transition temperature (temperature below which a polymer will lose elastomeric properties

50
and becomes hard and brittle like glass), sealants perform inadequately due to the stiffening

of the material and failure stresses may reach in the seal or at the adhesive interface as the

joint widens (Ketcham, 1995). Therefore, it was considered necessary to investigate the

effects of maximum and minimum service temperature exposure on sealant performance.

Tension test specimens were used for this test. Two sets of sealant samples, each

containing 3 foam sealant and 3 solid sealant specimens, were prepared. After a 2-week cure

at room temperature, one set of samples was exposed to uniform high temperature of 70ο C

and another set was subjected to uniform low temperature of -36ο C conditioning for a week

period. These samples were then subjected to tension tests and the changes in its

extensibility, stiffness and bonding characteristics were recorded.

4.9.2 Freeze and Thaw Cycle Test

Two sets of sealant samples, one containing 3 foam and another containing 3 solid

tension test specimens, were prepared. After a 2-week cure at room temperature, both set of

samples were subjected to alternate cycles of low temperature (-29οC) and room temperature

(+24οC) conditions for a week period. Finally, all specimens were pulled to the failure in

Instron machine at a uniform crosshead speed of 10mm/min. The modulus and extensibility

of sealants were then compared with those obtained from tension test.

4.10 CURE RATE TEST

Sealant should cure fast enough to accommodate typical daily thermal movements

and differential joint movement caused by traffic without being damaged. Development of

sufficient integrity and strength in small time period is essential in maintenance work, such

51
as bridge joint resealing, to minimize traffic disruption. Tension test specimens were used to

determine the cure rate of each sealant.

The test procedure involved repeated elongation of sealant specimens to 100 % strain

at different time intervals and recording the corresponding stress values. For the first 24 h

after the specimens’ preparation, the test was performed at 3, 6, 18, and 24 h, after which the

test was repeated at every 24 h period for 42 days.

4.11 TACK FREE TIME TEST

Tack free time is important in characterizing sealant as it is a measure of the surface

cure time and sealant’s resistance to dirt pick-up and impinging rainfall. It is the time

required for the sealant surface to be non-sticky and usually determined by visual

observation. Here, it should be understood that sealant cure reaction continues at molecular

level for much longer time after being visibly completed and therefore, sealants may not have

developed sufficient structural integrity to sustain large deformations after the specified tack

time. For joint sealants, tack free time is desired to be as small as possible such as to cause

minimal disruption to the road traffic. ASTM C 679 Test for tack free time was used in the

laboratory. The test involved lightly touching a surface of a curing sealant with a

polyethylene film at the regular intervals until sealant did not attach itself to the film.

4.12 WATER TIGHTNESS TEST

One of the critical requirements for the successful performance of any joint sealant is

that it provides an effective barrier preventing water from passing through the joint on to the

substructure elements below. The simple laboratory test to check watertight integrity of

52
sealant consisted of a hollow glass cylinder of 38.1 mm (1.5 in.) diameter and 63.5 mm (2.5

in.) height. The sealant was poured in the cylinder such as to form a sealant ring of 12.7 mm

(0.5 in.) thickness at one end of cylinder which bonded well with the inner surface of the

cylinder. A column of water, 38.1 mm (1.5 in.) high, was allowed to pond above the sealant

for a period of 96 h and observation was made for any water leakage.

4.13 TESTS ON WEATHERED SEALANT

Deterioration in physical properties can occur when the elastomeric joint sealants are

exposed to the weather effects (rain, snow, wind, and sunlight). This deterioration can be

observed as cracking, peeling, chalking, color changes, and other surface defects, and

ultimately by failure of the material. Presence of ozone, sunlight, oxygen, moisture, and

temperature are the cause of deterioration. UV light present in sun radiation can cause

surface hardening which can lead to crazing, chalking, and gradual erosion of the surface of

the elastomer. Oxygen can cause oxidation of the polymer and leads to the loss of mechanical

properties and elasticity and this effect is accelerated at high temperatures. Oxygen contained

in dew can also cause internal oxidation.

Three tension test specimens and two shear test specimens of each sealant type (foam

and solid) were prepared. Sealant specimens were exposed to outdoor condition in

Connecticut beginning immediately after their preparation in laboratory for a period of nearly

six and one-half months starting from the 14th of December, 2004 to the 28th of June, 2005.

The intent was to subject sealants to environmental effects during both cold and hot weather

seasons. Such outdoor exposure is expected to simulate the real field conditions for a bridge

joint sealant with the absence of vehicular effects being the only major exception. One of the

53
bases for evaluation was the change in the appearance due to surface degradation, dirt

pickup, and mildew. Besides, a variety of tests including tension, shear, stress relaxation, and

loading and unloading were performed on the weathered sealant samples to determine

outdoor exposure effects on their properties.

54
 5. LABORATORY TEST RESULTS AND DISCUSSION

Results from a wide variety of laboratory tests outlined in chapter 5 are presented and

discussed below.

5.1 TENSION TEST RESULTS

5.1.1 Loading Test Results

The following properties were evaluated from the tensile test: ultimate tensile

strength (kPa), ultimate elongation (%), modulus (stress) at 100 % strain, loading-unloading

behavior, and failure mode (adhesive/cohesive). Ultimate values refer to the values at the

point of maximum load where failure initiated. The nominal or engineering stress and strain

were calculated considering the original unstrained cross sectional area (12.7 mm x 50.8 mm)

and original unstrained length (12.7 mm) of the sealant. Strictly, the stress should be the

force per unit area of the actual deformed section (true stress) but this is rather more difficult

to calculate. The modulus at 100 % strain was taken as the stress to extend a standard

specimen by the amount same as its original length. This is due to the fact that a stress/strain

curve for rubber does not contain linear elastic portion as is usual with, for example, metals,

and hence, modulus is not measured as such but quoted as the stress at some percentage

elongation.

Figure 29 shows typical stress-strain behavior of the two sealant types (foam sealant

and solid sealant). It can be seen that stress–strain dependencies are not linear in nature, and

therefore, Hooke’s law, in general, is inapplicable. Stress-strain curves also illustrate low

resistance and high extension responses for the sealants.

55
 F1, F2, F3 – Foam specimens
W1, W2, W3 – Solid specimens

Figure 29. Tensile behavior of Foam sealant and Solid sealant.

Table 4 summarizes the results from the tests performed on these sealants. The

nominal ultimate strains for the three samples tested range from 597 % to 608 % for the foam

sealant and from 374% to 607% for the solid sealant. The ultimate stress for the foam sealant

varies from 80 kPa to 103 kPa and for solid sealant from 186 kPa to 251 kPa. It was observed

that both sealants followed the similar non linear stress-strain trend. However, they showed a

significant difference in the magnitude of the tensile stress developed. The average 100%

modulus value (25 kPa) of three foam sealant specimens is found to be nearly 3 times less

than that of the solid sealant (76 kPa). The lower stresses are important for maintaining the

bond between the sealant and the concrete; at a given gap displacement, the foam-concrete

interface will experience about 1/3 of the stress as that of the solid sealant. Cohesive mode

56
(failure within material) was observed to be the dominating failure mode for both types of

sealants during the tension test.

Table 4. Summary of Stress, Strain and Failure Mode Results from Tension Tests
Sealant Specimen Ultimate Ultimate Modulus @ Failure
Designation Nominal Nominal 100 % Strain Mode
Strain, (%) Stress, (kPa) (kPa)
Foam F1 597 103 26 Mixed
Silicone F2 608 94 24 Cohesive
F3 604 80 24 Cohesive
a
Average 603 ± 13 92 ± 30 25 ± 3 -
Solid W1 444 210 79 Cohesive
Sealant W2 374 186 80 Cohesive
W3 607 251 69 Mixed
Average 475 ± 296 216 ± 81 76 ± 14 -
a
95% confidence interval for the average.

5.1.2 Loading and Unloading Test Results

Figure 30 shows loading and unloading behavior of sealants in simple tension for five

consecutive cycles. Upon unloading the sample attains its original undeformed shape. The

observed elasticity is entropy-driven in nature and elastic force is due to changes in the

conformational entropy (James and Guth, 1942). The long chain molecules are stretched out

to statistically less favorable states which gives rise to an elastic recovering force that tends

to reduce the value of end to end distance of chains to that corresponding to maximum

entropy which corresponds to undeformed state. There are also some hysteresis effects. The

hysteresis was found to be more significant in the first loading-unloading cycle. This

phenomenon, common to most rubbers (Medalia and Kraus, 1994), may be attributed to the

delayed relaxation of chain ends that occurs when the sample is deformed. In foams, an

additional process of gas diffusion in and out of the cells can add to the hysteresis.

57
 Figure 30. Tension loading and unloading.

It is observed that when a sealant specimen is stretched after the first cycle, the

second stress–strain curve lies below the first one. The phenomenon is called the stress

softening which may be attributed to more than one mechanism. One of these is simply

incomplete elastic recovery (Mullins, 1969). Another mechanism is the progressive

detachment, or breaking of network chains attached to filler particles (Bueche, 1960). (All

silicones are filled with silica particles.) According to this mechanism, known as the Mullins

effect, during the first extension, chains pull free from the filler particles. In the second and

subsequent extensions, these chains are no longer supporting stress and the sealant is

softened. In our tests, the stress-strain curves followed almost the same path as the one

corresponding to the 2nd extension in agreement with the Mullins hypothesis. The hysteresis

58
effect is observed to be more pronounced for the solid sealant than the foam sealant, but the

stresses levels are also higher.

5.2 COMPRESSION TEST RESULTS

5.2.1 Loading Test Results

Figure 31 shows the compression behavior for the two types of sealants (foam sealant

and solid sealant). As expected, it was observed that the compressive stresses developed in

the solid sealant were much higher than those in the foam sealant. The average compression

moduli at 50 % strain were 26 and 193 kPa for the foam and solid sealants, respectively.

5.2.2 Loading and Unloading Test Results

Figure 32 shows sealant behavior when they were repeatedly loaded and unloaded in

compression. Hysteresis effects are observed in case of both sealants with no residual stress

being developed when the sealant reached their original uncompressed state (zero strain).

The hysteresis effect is observed to be smaller in case of the foam sealant compared to the

solid sealant.

59
F1, F2, F3 – Foam specimens
W1, W2, W3 – Solid specimens

Figure 31. Compression behavior of sealants.

Figure 32. Compression loading and unloading.

60
5.3 SHEAR TEST RESULTS

Figure 33 shows the stress strain behavior of each sealant type under simple shear.

Table 5 presents the test results. For rubber elastomers subjected to shear deformation, the

theory of linear relationship of stress and strain (neo-Hookean behavior) is in good

agreement with the experimental results up to a shear strain (γ) of about 1.0 (Treloar, 1970).

The average elastic shear modulus of the foam sealant was found to be 11.43 kPa. The

corresponding value for the solid sealant was 23.42 kPa. For shear strain values greater than

1.0, the sealants were no longer observed to remain under simple shear but are subjected to a

more complex deformation due to the finite size of the sample.

F1, F2, F3 – Foam specimens

W1, W2, W3 – Solid specimens

Figure 33. Simple shear test results.

61
 Table 5. Results of Shear Test
Sealant Specimen Shear Ultimate Ultimate Force, Failure
Designation Modulus, Deformation, (N) Mode
(kPa) (mm)
Foam F1 12 55 302 Cohesive
Sealant F2 11 48 261 Cohesive
F3 12 43 287 Cohesive
Wabo W1 24 37 363 Adhesive
Seal W2 22 44 401 Mixed
W3 23 40 367 Mixed

5.4 SALT WATER IMMERSION TEST RESULTS

Figure 34 presents the results of tension test on salt water immersed and dry foam

sealant specimens showing effects on the two performance indices modulus and extensibility.

Similar results for the solid sealant are presented in Fig. 35. Figure 36 shows comparison

between the tensile properties of these two sealant types after immersion in salt-water

solution for 2 weeks.

F1, F2, F3 – F4, F5, F6 –

Immersed specimens Dry specimens

Figure 34. Effects of salt water immersion on foam sealant.

62
 W1, W2, W3 - Immersed specimens
W4, W5, W6 – Dry specimens

Figure 35. Effects of salt water immersion on solid sealant.

F1, F2, F3 – W1, W2, W3 –

Foam specimens Solid specimens

Figure 36. Comparison of effects of salt water on foam sealant and solid sealant.

63
 Table 6 presents the results of tension tests on salt water immersed samples of both

foam sealant and solid sealant. Table 7 compares mean 100% modulus and mean ultimate

elongation values of salt-water-immersed samples with those obtained from the tension test

on dry samples (Table 4 and Figure 30). Both dry and immersed samples were prepared at

the same time and hence the results for these two samples have been compared assuming

equal variances. It can be observed from the results presented in Fig. 34 and Table 7 that

there is decrease in the extensibility of foam sealant with immersion, from the average

ultimate elongation value of 603% (for dry specimen) to the value of 453% (salt water

immersed specimen). A two-tail t-test was performed to determine if this observed difference

in the mean extensibility of immersed and dry samples was a chance finding or not. The test

resulted in a t-value of 7.9 and a probability of 0.0014. Thus, the extensibility difference

between two groups was not likely to have been a chance finding and they are statistically

different. The modulus of foam sealant at 100% strain was found to have increased by about

15% after immersion. The reason for this might be that the curing of the specimen was

enhanced after immersion (Beech, 1988). With enhanced curing, density of crosslinking

increases which is manifested by increase in the measured modulus. The comparison of

failure modes observed for dry and immersed foam sealant specimens did not show any

evidence of bond loss between the sealant and concrete substrate even at the ultimate

elongation.

For the solid sealant, however, the effects of the immersion appeared to weaken its

bonding characteristic. As can be seen in Figure 35 and Table 6, two out of three specimens

(W2 and W3) showed total adhesion failure (complete separation of sealant from substrate)

during the tensile test. These failures occurred at relatively small strain values of 191%

64
(specimen W3) and 71% (specimen W2). The third specimen (W1) showed the mixed failure

(partly cohesive and partly adhesive) at 366% elongation value. The diffusion of salt water

along the sealant-substrate interface is a possible reason for the weakening of the sealant

bond to the joint surface (Beech, 1988). The measured values showed an apparent 10%

decrease in the 100% modulus of the solid sealant due to immersion, but this change was not

statistically significant. It should be noted here that the sealant joints were prepared without

using primers. Primers may affect the bonding of the foam and solid sealants with the

substrate by different amount, and hence could yield different bonding test results than

reported herein.

Table 6. Stress, Strain and Failure Mode Results from Salt Water Immersion Test
Sealant Test Specimen Strain Stress 100 % Failure
Condition Designation (%) (kPa) Modulus Mode
(kPa)
Foam F1 437 98 27.7 Mixed
Silicone Immersed F2 491 98 29.3 Cohesive
F3 433 86 28.7 Cohesive
Average 453 ± 81a 94 ± 17 28.5 ± 2.0 -
Solid W1 366 180 76 Mixed
sealant Immersed W2 71 53 No result Adhesive
W3 191 93 61 Adhesive
Average 209 ± 368 108 ± 168 - -
a
95% confidence interval for the average.

Table 7. Salt Water Immersion Effects on Sealant Modulus and Extensibility

Sealant Test Average Average Average Extensibility Modulus
Condition Ultimate Ultimate 100 % Difference Difference
Strain (%) Stress Modulus (%) * (%) *
(kPa) (kPa)
Foam Dry 603 92 24.8 0 0
Silicone Immersed 453 94 28.6 -25 +15
Solid Dry 475 216 76.1 0 0
sealant Immersed 209 109 68.8 -56 -10
* Note: + indicates increase and – indicates decrease from dry specimen tension test results

65
5.5 BOND (OVEN AGED) TEST RESULTS

Upon visual observation, no specimens were found to develop any crack, separation,

or other opening in the sealant or between the sealant and the concrete test blocks. For each

of the five extensions, 100 % modulus (the sealants stress at 100% strain) values have been

plotted in Fig. 37, which depicts the fatigue of material with tensile deformation after oven

aging and cooling cycles. These values are summarized in Table 8. From these values, it is

seen that the 100% modulus of both sealant types changed by small amounts with each

subsequent extension in the course of five extensions. Figure 38 shows the stress-strain plot

for the 5th extension of the sealants. This plot represents a typical of stress-strain curve

observed during each extension. Comparison of mean 100% modulus values between the

first and the fifth extension shows a modulus decrease of 13% and 7% for the foam and the

solid, respectively. Both of these changes were not statistically significant. However, the

trend with cycle for the foam has a t value of 3.9 for a probability of 0.05 at 3 degrees of

freedom. For the solid, the probability level is 0.17, which we do not consider significant.

Table 8. Modulus under Repeated Extensions from Bond Test after Oven Aging
Sealant Specimen Modulus @100 % Strain (kPa)
Designation
1st 2nd 3rd 4th 5th
Foam F1 Extension Extension Extension Extension Extension
Silicone F2 35 33 31 31 31
F3 33 31 30 29 29
28 27 23 24 25
a
Average 32 ± 9.0 30 ± 7.0 28 ± 11.0 28 ± 9.0 28 ± 8.0
Solid W1 88 82 81 79 83
sealant W2 88 81 80 78 79
W3 70 66 65 68 67
Average 81 ± 25.0 76 ± 21.0 75 ± 22.0 75 ± 15.0 76 ±20.0
a
95% confidence interval for the average.

66
 100
100% modulus, kPa Bond test after oven aging

80 Solid

C-T equation
60

40
Foam
20
1 2 3 4 5 6
Cycle number
Figure 37. Fatigue of sealants with tensile deformation after oven aging
and cooling cycles. Curves are simplified C-T equation

F1, F2, F3 – Foam specimens

W1, W2, W3 – Solid specimens

Figure 38. Bond Test: Stress-Strain at 5th extension.

67
 In Fig. 37, C-T equation (Eq. 8) has been used to describe the loss of sealant modulus

with tensile deformation after oven aging and cooling cycles. The parameters of the model

equation were calculated by nonlinear regression analysis using commercially available

software SigmaPlot 9.0 (Systat Software Inc., 2004). The value of m was set at 1.0. The

values of E∞ are 26.7 kPa and 73.3 kPa for the foam and solid sealant respectively. The

values of a , which represents the decay “time” (number of cycles) are 0.21±0.09 and

0.11±0.07 for the foam and solid respectively, where the errors are the 95% confidence limits

of the values. Comparing these two values using Student’s t-test gives a t-value of 2.8 and a

probability of 0.03. Thus the two rates of decay are different at the 97% probability level.

The foam sealant, therefore, appears to be more resistant to fatigue than the solid sealant,

based on the available data. Our interpretation of this is that the lower modulus of the foam

results in less stress at the fixed strain, and thus less fatigue.

5.6 STRESS RELAXATION TEST RESULT

Figures 39 and 40 respectively show stress relaxation test results in tension and

compression that have been reduced by Eq. 10 to obtain relaxation times. During stress

relaxation in tension, stress decreased from the initial value of 27.62 kPa to 20.91 kPa for the

foam and from the initial value of 92.68 kPa to 82.76 kPa for the solid in 24 hour period.

Hence, relaxations of two sealants are calculated using Eq. 9 as 25% and 10% respectively

after total test duration of 24 h. A sharp decrease in the stress was observed to take place

within small initial test period of 1 hour beyond which rate of stress decay was gradual (Fig.

39). For example, for the foam, stress drop within initial 1 hour was 4.4 kPa whereas the drop

was 2.31 kPa for the subsequent 23 hours period. For the solid, corresponding drops were

68
6.44 kPa and 3.48 kPa respectively. It was also observed that after the 16-hour period, the

stresses appeared to remain more or less steady for both sealants depicting the achievement

of equilibrium. During stress relaxation in compression, stress dropped from the initial value

of 70.00 kPa to 19.83 kPa for foam and from the initial value of 254.1 kPa to 227.7 kPa for

solid in the test period of 24 hours. Corresponding relaxations are found to be 70% and 10%

from Eq. 9. As observed in stress relaxation in tension, the majority of stress decay in

compression also occurred in initial small period after which the rate of decay was gradual.

For the foam, the stress decay during the first 1 hour was 23.64 kPa, and the decay during

next 23-hour test period was 26.53 kPa. Similarly, corresponding stress drops for the solid

were 19.4 kPa and 7.0 kPa.

The stretched exponential form of the relaxation modulus E(t) given by Eq. 10 was

used in the Fig. 39 and Fig. 40 to model the isothermal stress relaxation behavior of sealants.

The parameters for foam and solid sealants for different relaxation test modes (tension and

compression) have been presented in Table 9. These parameters were calculated according to

the Eq. 10 using SigmaPlot 9.0 graphing software. The values of relaxation constant τ in

tension test are 0.91± 0.02 h and 0.93±0.02 h for the foam and solid, respectively, where the

errors are the 95% confidence limits of the values. The t-test for these two values gives a t-

value of 0.4 which indicates that the difference between these two values is not significant.

Table 9. KWW Equation Parameters for Stress Relaxation Test

Sealant Relaxation β E0, kPa E∞, kPa τ, h
Test Mode
Foam Tension 0.3 28.7 20.4 0.91±0.02
Silicone Compression 0.55 75.8 20.8 2.2±0.04
Solid Tension 0.36 93.5 82.2 0.93±0.02
sealant Compression 0.17 254.6 208.6 41.4±0.77

69
 Therefore, two sealants are found to exhibit approximately same decay rate in tensile

elongation. For compression relaxation, however, large difference between relaxation times

is observed. The values of relaxation constant τ are 2.2±0.04 h and 41.4±0.77 h for the

foam and solid, respectively in this case. Comparing these two values gives a t-value of

3.04. Thus, two relaxation time values are statistically different and it appears that the

stress in the foam sealant gets relaxed much faster than the stress in the solid sealant. It

should be noted herein that because the relaxation tests were performed in tension and

compression mode rather than shear mode, there might have been the gain or loss of air in

the foam sealant bubbles during prolonged exposure of the specimen to constant strain

condition. As a result, the relaxation behavior observed for the foam sealant in our case

might not be due purely to the material response alone.

70
Figure 39. Isothermal stress relaxation in tension for foam and solid. Curves
are KWW equation.

71
Figure 40. Isothermal stress relaxation in compression for foam and solid.
Curves are KWW equation.

72
5.7 CREEP TEST RESULTS

Deformation of foam and solid with time under constant tensile load are presented in

Fig. 41. Creep for foam and solid after 24 hours were calculated using Eq. 11 as 21% and

15%, respectively.

Initial Deformation
D1 (Foam) = 10.18 mm
D1 (Solid) = 9.78 mm

Figure 41. Creep deformation of foam and solid in tension.

5.8 COMPRESSION RECOVERY TEST RESULTS

Figure 42 shows the test results after the sealants have been released from the

compressed state. Foam sealant was observed to undergo a permanent set of nearly 20% of

initial compression while solid sealant showed as little as 1.5% set. The percent value in case

of the foam sealant seems to be substantially larger than that for the solid sealant, however; it

was the value observed just 1 hour after the release of sealant from compression. During the

73
prolonged compression of the foam, air is forced to escape from the foam bubbles. Upon

release of compression, there is a backflow of air into these bubbles from the surrounding

and it may take significantly longer time than just 1 hour to complete the process of the

regaining of air by the foam bubbles. Hence, the set observed for the foam sealant may not be

the permanent in nature. Recovery attained by the foam sealant is about 90% of its original

width/length, whereas, it is almost 100% for the solid sealant. It is also observed that for both

sealants, almost all recovery takes place within the first five minutes after the sealants have

been released from compression.

Solid

Foam

Thickness after Initial

compression:
ts (Foam) = 12.7 mm
ts (Solid) = 12.7 mm

Figure 42. Compression recovery of sealants.

5. 9 TEMPERATURE SENSITIVITY TEST RESULTS

5.9.1 High and Low Temperature Effects Test Results

The effects of exposing sealants to low and high temperature extremities on their

important tensile properties, namely ultimate elongation and modulus, and their bonding

74
integrity with the joint substrate have been presented herein. Figure 43 shows stress-strain

curves from the tension test on foam and solid samples which have been subjected to high

temperature (70 ºC) exposure. Similar results for samples exposed to low temperature (-36

ºC) are presented in Fig. 44. Numerical values of test results are summarized in Table 10.

These test results are also compared with results obtained from tension test that were

performed on the same aged foam and the solid sealant samples cured at standard laboratory

conditions (23 ± 2° C).

Table 10. Stress, Strain and Failure Mode Results from Temperature Sensitivity Test
Sealant Exposure Specimen Ultimate Ultimate Modulus @ Failure
condition Designation Nominal Nominal 100% Mode
Strain, (%) Stress, (kPa) Strain,(kPa)
High F1 350 87 32 Cohesive
Temp F2 306 64 30 Adhesive
Foam F3 275 65 31 Cohesive
Silicone Average 310±94a 72±33 31±3.0
Low F1 204 45 26 Adhesive
Temp F2 457 78 22 Cohesive
F3 427 51 19 Adhesive
Average 363±343 58±44 22.5±8.0
High W1 335 207 93 Adhesive
Temp W2 240 161 91 Adhesive
Solid W3 225 150 92 Cohesive
Sealant Average 267±148 173±76 92±4.0
Low W1 208 93 59 Adhesive
Temp W2 472 182 65 Cohesive
W3 640 206 60 Adhesive
Average 440±541 160±148 61±8.0
a
95% confidence interval for the average.

Table 11 presents mean 100% modulus and mean ultimate elongation values for

standard temperature conditioned (Table 4 and Fig. 29) and high and low temperature

conditioned samples (Table 10 and Fig. 43 & 44).

75
Figure 43. Stress-Strain plot of samples after high temperature conditioning.

Figure 44. Stress-Strain plot of samples after cold temperature conditioning.

76
 A multivariant linear model (Eq. 13) used in Fig. 45 to model the experimental

modulus data which included temperature and material effects as well as effects due to

interaction of material with temperature was:

E = a0+a1*X1+a2*T+a3*T*X1 --------- (13)

Where, E = modulus, T = temperature, °C; X1= 1 if foam, zero otherwise.

All parameters ai were found to be significantly different from zero and are,

respectively: 71±5; -46±7; 0.28±0.1; and 0.21±0.16. Figure 45 shows the results of

multivariant linear model for material effects and interaction of material with temperature

which suggests that high temperature conditioning of the sealant increases the modulus of

sealants. It is also observed that slope of the solid is greater than foam. The oven aging has

appeared to assist the curing process of sealants which become apparent in increase in

measured modulus of sealants. It was observed that majority of foam specimens failed

cohesively; however, adhesive failure mode was observed for most of solid specimens. The

stress-strain plot results (Fig. 29) from the tension test makes it clear that the solid sealant

fails at substantially higher levels of stress than the foam sealant. The higher stress levels

required to elongate the solid sealant result in very high interfacial stress at the

sealant/concrete interface which explains degradation of bond between solid sealant and joint

substrate. In contrast, the low modulus, and hence, lower stress levels at foam sealant-

concrete interface results in the material failing rather than debonding from the concrete.

77
 100
Aged seals
100% modulus, kPa 90
80 Solid
70
60
50
40
Foam
30
20
-40 -20 0 20 40 60 80
o
Temperature, C
Figure 45. Effects of high and low temperature exposure on sealant modulus. The
lines are results of multivariant linear model for material effects and interaction of
material with temperature.

Figure 46. Effects of high and low temperature exposure on sealant extensibility.

78
 Table 11. Temperature Exposure Effects on Sealant Modulus and Extensibility
Sealant Conditioning Average Average 100% Extensibility Modulus
Temperature Ultimate Modulus, Difference, Difference,
Strain, (%) (kPa) (%) * (%) *
Foam Standard (23ºC) 603 24.80 0 0
Silicone
High (70 ºC) 310 31.00 -49 +25
Low (-36ºC) 362.7 22.5 -40 -9
Standard (23ºC) 475 76.13 0 0
Solid High (70ºC) 267 92.05 -44 +21
Sealant
Low (-36ºC) 440 61.4 -7.3 -19
* Note: + indicates increase and – indicates decrease from standard specimen test results

For sealant elongation, the parameters were not found to be significant suggesting

that the material and temperature effects were negligible on sealant elongation. However, a

drop in ultimate displacement for two sealants under cold and hot exposure condition was

noted (Fig. 46). The magnitudes of ultimate strain drops for low and high temperature

conditioned samples compared to standard samples are given in Table 11. The reason for

strain reduction at cold temperature is due to the early sealant and substrate bond failure

which might be attributed to the diffusion of moisture into the joint interface during cold

exposure. Also, maximum percent strain reduction of oven aged sealants might be ascribed to

the stiffening of the sealants after high temperature exposure.

5.9.2 Freeze and Thaw Cycle Test Result

The variation of conditioning temperature during one typical cycle of duration 24

hours is shown in Fig. 47. Results of tension test on the foam and solid specimens are

presented in Fig. 48 and Table 12.

79
 Figure 47. Temperature variation in a single cycle.

Figure 48. Tension test results of sealants exposed to alternate

cold and room temperature.

80
 When we compare them with the results obtained from the tension tests on standard

temperature conditioned sealant samples, we see maximum strain percent decrease of 24%

and 34% for foam and solid respectively. Modulus of solid sealant is not affected. However,

foam sealant is observed to be softened as there is 7% decrease in measured modulus.

Separation of sealant from the joint surface is seen to be the most common failure mode

among specimens during the tension test.

Table 12. Stress, Strain and Failure Mode Results from Freeze and Thaw Cycle Test
Sealant Specimen Ultimate Ultimate Modulus @ Failure
Designation Nominal Nominal 100 % Strain Mode
Strain, (%) Stress, (kPa) (kPa)
Foam F1 420 64 22 Adhesive
Silicone F2 500 65 27 Cohesive
F3 449 60 20 Adhesive
Average 456±101a 63±7.0 23±9.0
Solid W1 323 170 78 Adhesive
Sealant W2 327 161 75 Cohesive
W3 285 153 79 Adhesive
Average 312±58 161±21 77±5.0
a
95% confidence interval for the average.

5.10 CURE RATE TEST RESULTS

Figure 49 shows the sealant stress as the function of curing time. Table 13 presents

the 100% modulus values achieved by sealants at different curing periods. The results show

higher increase in sealant strengths with time up to the period of approximately 3 days which

suggests rapid curing of sealant in the initial period. During this initial 3-day period, sealants

were observed to develop nearly 64% of their 21-day strength, a value that has been assumed

to be the full-cured sealant strength. The increments in sealant strength were small after this

period indicating the more gradual curing of the sealants afterwards. It was observed that

81
80% or more of the 21-day sealant strength was developed within the initial 7 days period of

curing.

Solid

Foam

Figure 49. Curing rate of foam and solid.

Table 13. Curing Rate of Sealants

Sealant 100% Modulus, kPa
3 days 7 days 14 days 21 days 42 days
Foam 16.8 21.4 25.8 27.2 30.4
Silicone
Solid sealant 56.9 76.3 87.5 89.1 95.9

5.11 TACK FREE TIME AND WATER TIGHTNESS TEST RESULTS

Tack free time for the foam sealant and solid sealant were found to be approximately

50-80 minutes and 30-60 minutes respectively. During Water tightness test, the non-

82
submerged bottom surface of each sealant was inspected for any evidence of dripping water

or moisture. No sign of water leakage was detected in both sealants throughout the 96-hour

test duration

5.12 RESULTS OF TESTS ON WEATHERED SEALANT

The distribution of maximum and minimum temperature to which sealants were

exposed for a period of more than six months is shown in Figure 50. Similarly Fig. 51 and

Fig. 52 show the precipitation and rainfall amount during the exposure. The exposure began

on the14th of December, 2004 and ended on the 28th June, 2005. The day climate data

including maximum and minimum temperature, rainfall, and snowfall for the total test

duration presented herein were obtained from (National Weather Service, 2005).

Figure 50. Maximum and minimum temperature distribution.

83
 Figure 51. Cumulative precipitation.

Figure 52. Cumulative snowfall.

Based on visual observation, samples were checked for crazing (fine cracks),

chalking, dirt pick up, and surface erosion. Both sealants were found to exhibit none of these

degradations. It should be noted that no samples were kept at dark in room condition for

similar period as weathered samples to serve as the control.

84
5.12.1 Tension Test Result

Figure 53 is the stress-strain curve for weathered sealants. Increase in sealant stiffness

at the cost of maximum percent strain reduction is observed. The 100% modulus of foam and

solid sealants were measured to be 47 kPa and 101 kPa respectively. Elongations at break

were 263% for foam and 344% for solid. Comparison of these results with tension test results

of regular laboratory samples (3 week cured) shows increase in foam and solid stiffness by

90% and 75% respectively. Similarly, drops in elongation capacity were found to be 56% and

28% for foam and solid respectively. The observed stiffening of the sealant may be attributed

to oxidation of polymers which results in loss in mechanical properties and elasticity. The

continued curing reaction may have added to the stiffening of sealants by oxidation. The

curing reaction normally continues at molecular level for much time even after it visually

appears to be complete. The effects of oxidation and continued curing process in stiffening of

sealant are however difficult to separate out as there were no control samples (kept at dark

and room condition for similar period as weathered samples) to make a comparison.

Figure 54 shows the loading and unloading behavior of weathered sealants in simple

tension. The strain limit was set to 200%. Only one loading and unloading cycle could be

achieved for the solid sealant as a part of concrete block bonded with sealant chipped off

causing tearing apart of sealant during the second cycle loading of specimen. The behavior is

the same as the one observed in regular tension test described in section 6.1, only difference

being that the higher level of stresses were observed this time indicating the stiffening of the

sealants.

85
 Figure 53. Tension test results: weathered sealants.

Figure 54. Tension loading and unloading of weathered sealants.

86
5.12.2 Shear Test Results

Results of shear test performed on weathered foam and solid sealant are presented in

Fig. 55. Average elastic shear modulus of two sealants are 21 kPa and 43 kPa respectively.

The corresponding values obtained for regular 2-week cured laboratory conditioned samples

were 11.4 kPa and 23.4 kPa (see section 5.3). As observed in tension test of weathered

samples, stiffening of sealants is observed in this case also and similar reasoning could be

argued for this observed behavior.

Figure 55. Shear test results of weathered sealants.

5.12.3 Stress Relaxation Test Results

Figure 56 and 57 respectively show stress relaxation test result for weathered foam

and solid sealants. The decay of stress with time was recorded for 96 hour period. The total

87
relaxations for the foam and the solid sealants during the test period are 20% and 14%

respectively (Eq. 9). A Kohlrausch-Williams-Watts stretched exponential function (Eq. 10;

repeated here for simplicity) was applied to the sealant stress relaxation.

E(t) = (E0 - E∞) exp[-(t/τ)β] + E∞ ----------(10)

Parameters β, E0, and E∞ for foam and solid sealants are presented in Table 13. The values of

relaxation constant τ in tension test are 1.5±0.004 h and 1.5±0.002 h for the foam and solid,

respectively, where the errors are the 95% confidence limits of the values. The t-test for these

two values gives a t-value of zero which indicates that the difference between these two

values is not significant (critical value of t is 1.96 at p = 0.05). Therefore, two sealants are

found to exhibit indistinguishable decay rates in tensile elongation. It was also observed that

experimental relaxation curves slightly oscillate about the theoretical exponential stress

decay curves.

Table 14. KWW Equation Parameters for Stress Relaxation Test: Weathered sealants
Sealant β Eº, kPa E∞, kPa τ, h
Foam Silicone 0.23 51 42 1.5±0.004
Solid sealant 0.24 108 92 1.5±0.002

88
 KWW equation

Figure 56. Isothermal stress relaxation in tension. Weathered foam sealant.

KWW equation

Figure 57. Isothermal stress relaxation in tension. Weathered solid sealant

89
 6. CONCLUSIONS

The main purpose of this study was to develop and evaluate an alternate low cost and

effective sealing material for small movement bridge expansion joints. A silicone foam

sealant was developed in the laboratory as a potential joint sealing candidate by modifying

one of the commercially available bridge joint sealants, Wabo silicone seal (Watson Bowman

Acme Corp., 2003). Other commercial sealants such as Dow Corning RCS 902 (Dow

Corning Corporation, 2004a) with composition similar to that of Wabo silicone seal are

available in the industry. However, due to the limited resources, only one of the commercial

sealants was chosen for this research. A series of laboratory tests was performed on the

newly developed silicone foam sealant to evaluate its material and mechanical

characteristics. In addition, for comparison purpose commercial Wabo silicone seal (solid

sealant) was also assessed in the laboratory. The following conclusions can be drawn based

on the results obtained from the present study:

1. The foam sealant was observed to undergo a significant rise (70%) in its volume on

curing. This is a desirable feature in that it can lead to the saving in the cost of the joint

sealing material and a more certain filling of cavities in the joint.

2. The foam sealant was found to have lower stiffness and greater extensibility than the

solid sealant. Low modulus is desirable and is particularly important because it will help

reduce cohesive and adhesive stresses developed in the seal during the joint operation.

3. When immersed in salt water solution at 45°C, the foam sealant appeared to lose its

ultimate elongation capacity from 603% (dry condition) to 453% but there was no bond

loss with the concrete substrate. Whereas it was observed that the salt water immersion

appeared to weaken the bonding between the solid sealant and the concrete substrate

90
 causing the sealant to pull away from the joint surface at relatively smaller average

elongation value (209%) than that (475%) of its dry counterpart. It should be noted here

that the sealant joints used herein for testing were prepared without the application of

primers to the substrate surfaces. Primers may enhance the bonding of the foam and solid

sealants with the substrate by different amount, and hence could yield different bond test

results.

4. When subjected to oven aged bonding test as per ASTM D5893, no specimens were

observed to develop any crack, separation, or other opening in the sealant or between the

sealant and the concrete test blocks. The foam sealant was observed to be more resistant

to the fatigue due to alternate cycles of cold extension and self-recompression than the

solid sealant.

5. Relaxation phenomena in sealants were found to follow stretched exponential decay law.

Compression relaxation times obtained for the sealants suggested that stress in foam

sealant relaxes faster than that in solid sealant. For tension mode, relaxation amounts

were 25% (foam) and 10% (solid) of the sealant stress at 100% extension. Corresponding

statistics for relaxation in compression mode were 70% and 10% for foam and solid,

respectively.

6. Creep for foam and solid after 24 hours were found to be 21% and 15%, respectively.

7. Foam sealant was observed to undergo a permanent set of nearly 20% of initial

compression while solid sealant showed as little as 1.5% set. These values were observed

1 hour after the release of sealants from compression.

8. Exposure to high temperature condition increased modulus and reduced ultimate strain

capacities of sealants. The oven aging appeared to assist the curing process of sealants

91
 which became apparent in increase in measured modulus of sealants. Exposure to cold

temperature condition resulted in large decrease in maximum percent elongation of foam

but the final strain value thus obtained was still much higher than the required maximum

in the field. The measured modulus values suggested that low temperature conditions

slowed down the curing process.

9. After the freeze-thaw condition, the sealants were observed to have some decrease in

maximum tensile strain percent. There was only slight effect on the sealant moduli.

10. Sealants were observed to develop nearly 64 % of 21-day curing strength within initial 3-

day curing period. Also, 80 % or more of the 21-day curing strength was found to achieve

by sealants within initial 7-day curing period.

11. Tack free time for both sealants was found to be less than one and a half hour, which is

considered in favor of sealants as the traffic disruption time will be minimal during joint

sealing operation.

12. None of the sealants showed any sign of water leakage during the water tightness test.

13. No significant deterioration was observed in the sealants (cracking, surface erosion etc.)

subjected to weathering effects for 6 and one-half month period. Laboratory tests

suggested higher stiffness after subjecting to weathering. Tension test showed the drop in

elongation at break. The increased stiffness and decrease in elongation might be

attributed to the oxidation of polymer and continued process of curing of sealants.

14. From the stress relaxation test of the weathered sealants, both foam and solid sealants

were found to exhibit similar decay rate in tensile elongation. The experimental

relaxation curves are observed to slightly oscillate about the theoretical exponential stress

decay curves.

92
 7. RECOMMENDATIONS FOR FIELD DEMONSTRATION AND

MONITORING (PHASE II)

7.1 RECOMMENDATIONS TO PROCEED TO PHASE II

The results of laboratory tests performed so far on newly developed foam sealant are very

promising of its successful performance as an economic and effective bridge joint sealant

material. The rise in the foam volume once the curing reaction is complete can lead to a

significant economy in the bridge joint sealant operation while its low modulus nature will

prevent excessive interfacial stresses from being developed at the sealant substrate interface

thus helping the bond between the sealant and the joint surface to remain intact during the

opening of the joint. Debonding of the sealant from joint substrate has been reported as one

of the major problems associated with commercially available silicone sealant. Various

laboratory tests including salt water immersion test results have indicated the better bonding

properties of foam sealant compared to the solid sealant. Besides above, foam sealant’s rapid

curing, easy to install, self-leveling, and low modulus properties make it a very suitable

candidate for joint sealing operation. It is strongly recommended that the sealant material be

tested in the field by installing the joint seal in a real bridge and its performance be

monitored continuously. Therefore, it is concluded that Phase II of this project work is

highly warranted and the research teams strongly recommends for the implementation of

Demonstration and Monitoring Phase (Phase II).

93
7. 2 RECOMMENDED MAJOR TASKS FOR PHASE II

The major tasks recommended for Phase II of the project are presented below. The tasks

are grouped under 3 categories (a) Pre-Field Installation, (b) Field Installation, and (c) Post-

Field Installation and Monitoring.

7.2.1 Pre-Field Installation Tasks

1. It is important to have sealant performance evaluated under cyclic loading as the bridge

joint is constantly subjected to dynamic loading due to vehicular traffic. Deflection

controlled cyclic loading test should be performed in laboratory to simulate dynamic

deflection of joint sealants, namely normal deflection due to temperature and shear

deflection due to wheel loads.

2. The sealant bonding tests so far were done with the concrete substrate. Since concrete

surfaces have irregularities and the sealants normally stick better with these rough

surfaces. Therefore, adhesion/bonding tests to steel should be done in the extension.

3. Foams sealant developed in this study seems to give better results than the solid sealant

without the primers. However, more investigation is needed to compare the bonding

performance between these two sealants in the presence of primers.

4. In order to eliminate the possibility of air gain and loss by foam sealant during stress

relaxation in tension and compression modes, it is recommended that the relaxation test

be performed in shear mode. Dynamic Mechanical Analyzer (DMA) machine is capable

of performing the relaxation test in shear using the sandwich arrangement.

94
5. Since the concrete and steel will be the most frequently encountered joint substrate

material in new construction and repair works, water tightness of sealant should be tested

on specimens bonded to these substrate materials.

6. Puncture test of sealant is essential to evaluate its ability to withstand debris forced into

the joint by moving traffic. Poking objects or devices of various shapes can be pressed

against the sealant with appropriate load magnitudes to see the occurrence of any damage

to the sealant.

7. Effects of humidity on the behavior of the foam sealants had not been evaluated in this

first phase of the project. This should be evaluated in Phase II of the project.

8. So far, small batches of foam sealant have been produced in the laboratory to prepare test

specimens of small dimensions (e.g. 0.5” length/width and 0.5” thickness). Hence, for the

field installation purpose, a larger volume of the sealant needs to be prepared and applied

to joints of larger dimensions as found in real bridges. The quality and basic behavior of

the sealant materials thus prepared and applied should be assessed.

9. A proper sealant dispensing tool should be designed to facilitate the effective mixing of

foam sealant components and its easy installation in the bridge joints in the field. An

applicator gun should be designed such that it will able to hold and produce sufficient

sealant volume to facilitate the quick joint sealing operation.

10. It is essential that the sealant component mixing and joint sealing operation be conducted

in the ambient field outside laboratory environments before proceeding to the real bridge

application. These operations can be done in prototype joints between 2 concrete blocks

or two steel beams outdoor in the field environment. This will give confidence in the

95
 preparation of joints, sealant pouring and mixing and joint installation, in general. A

suitable method of vertical joint sealing method should also be established.

7.2.2 Field Installation Tasks

It is envisioned that the field installation procedure for the foam sealant can be similar to

that used for the 2-part, cold applied silicone sealants currently in use for bridge expansion

joint sealing (for example Watson Bowman Acme Corp., 2003; and Dow Corning Corp,

2004b). The tasks for the field installation of the sealant thus include, but not limited to, the

following:

1. Selection of bridge(s): Selection of one or more candidate bridges for the joint

installation and monitoring should be accomplished in coordination with NETC

project Technical Committee members who come from the transportation agencies of

each of the six states in the New England region. For the first application and

demonstration of the sealant, a low traffic volume bridge with expansion joint

needing repair should be selected. It is also recommended that the sealant be applied

only to a couple of feet length of joint first and its performance be monitored and

evaluated for some extended duration, before applying the sealant to the whole width

of the bridge cross-section.

2. Joint preparation: The joint interface must be free of dirt, coatings, oil, grease, rust

and other contaminants. For this purpose, sand blasting is recommended. Air blasting

of the joint after sandblasting is necessary to remove all debris in the joint and

surrounding area. To insure cleanliness of each joint interface should be wiped with a

clean rag.

96
 3. Backer Rod: A backer rod is required to prevent the flow of the sealant through the

joint during joint sealing operation and to ensure the proper shape and depth of the

sealant reservoir. A closed-cell, expanded polyethylene foam rod with approximately

25% larger in diameter than the joint gap is recommended.

4. Mixing of Sealant Components: Sealant components should be applied using proper

dispensing equipment. They should be mixed thoroughly and applied at the weather

condition (room temperature ~23oC and dry) when they produce the most effective

joint sealing characteristics.

5. Application of Foam Sealant: The foam sealant begins to rise in volume once it is

poured in the joint during the initial curing period. Therefore, the amount (depth) of

sealant to be poured into the joint must be carefully monitored so that when it is fully

installed, some recess from the riding surface is maintained to prevent the direct

contact of sealant to the vehicle tires.

6. Traffic control during joint preparation and installation: The tack free time of foam

sealant is approximately 50-80 minutes. Traffic closing and reopening should be

appropriately monitored.

7.2.3 Post- Field Installation and Monitoring Tasks:

A variety of tasks, including monitoring and inspection of the joint sealant, are

recommended after the sealant is installed in the bridge joints. They include, but not limited

to, the following:

1. Regular visual inspection/monitoring for any cracks, separation of sealant from the

joint substrate, and damage due to debris.

97
 2. Continuous recording of joint movement – amount of opening and closing

3. Continuous recording of temperature

4. Recording of weather elements, rain-fall, snow fall, and extreme weather events, such

as hurricane, flooding, etc.

5. Traffic data recording – traffic volume, speed, vehicle types, loads, etc.

Some special instrumentation might have to be installed to monitor and record data as

noted above, such as joint opening and closing and traffic data.

98
 8. REFERENCES

1. AASHTO (2004). AASHTO LRFD Bridge Design Specifications 2004. American

Association of State Highway and Transportation Officials, Washington D.C.

2. ASTM (1997). “Standard Specification for Cold Applied, Single Component,

Chemically Curing Silicone Joint Sealant for Portland Cement Concrete Pavements.
ASTM D 5893”, Annual Book of ASTM Standards, Vol.04.03, American Society for
Testing and Materials, West Conshohocken, PA.

3. Beech, J. C. (1988). “The Performance of Sealants under Wet Conditions,”

R.I.L.E.M. Symposium on Building Joint Sealants, Boras, Sweden.

4. Beech, J.C. and Mansfield, C. (1990). “The Water Resistance of Sealants for
Construction.” Building Sealants: Materials, Properties and Performance
ASTM STP 1069, (O’Connor, T.F., ed.), American Society for Testing and Materials,
Philadelphia, PA.

5. Benson, K. (1986). “Bridge Deck Expansion Joint Evaluation.” Report No.

FHWA/AR-86/006, Arkansas State Highway and Transportation Department, Little
Rock, AR.

6. Biel T.D. and Lee H. (1997). “Performance Study of Portland Cement Concrete
Pavement Joint Sealants.” J Transportation Engineering, 123 (5), 398-404.

7. Bramel, B.K., Dolan, C.W., Puckett, J.A., and Ksaibati, K. (2000). ‘Asphalt Plug
Joints Refined Material and Tests and Design Guidelines.” Transportation Research
1740, Paper No. 00-0414, 126-134.

8. Burke, M.P. (1989). “Bridge Deck Joints.” NCHRP Synthesis 141, Transportation
Research Board, Washington, D. C.

9. Bueche, F. (1960). “Molecular Basis for The Mullins Effect.” J Applied Polymer
Science, 4(10), 107-114.

10. Chang, L.M., and Lee, Y.J. (2002). “Evaluation of Performance of Bridge Deck
Expansion Joints.” J. Performance of Constructed Facilities, 16(1), 3-9.

11. Chen, W. and Duan, L. (2000). Bridge Engineering Handbook. CRC Press, NY.

12. Connecticut DOT (2003). “Deck Joints”, Bridge Design Manual, State of
Connecticut Department of Transportation Newington, CT, section 10.

99
13. Connor R. J. and Dexter R. J. (1999 ) “Fatigue Design of Modulur Bridge Expansion
Joints.” J Transportation Research Record 1688, Paper No. 99-1238, 124-130.

14. Dahir, S.H. and Mellott, D.B. (1985). “Bridge Deck Expansion Joints.” Report No.
FHWA/PA-85-023, Pennsylvania Department of Transportation, Harrisburg, PA.

15. Dow Corning Corporation (2004a). “Dow Corning 902 RCS Joint Sealant.” Form No.
62-181F-01, Dow Corning Corporation, Midland, MI. (Website:
http://www.downcorning.com/)

16. Dow Corning Corporation (2004b). “Installation Guidelines and Equipment

Recommendations - Dow Corning 902 RCS Joint Sealant.” Form No. 62-272E-01,
Dow Corning Corporation, Midland, MI. (Website: http://www.downcorning.com/)

17. Dubiel, S. V.; Griffith, G. W.; Long, C. L.; Baker, G. K.; Smith, Robert E., (1983)
“Determination of reactive components in silicone foams.” Analy. Chemistry, 55(9),
1533-1537.

18. Eacker, M. J. and Bennett, A. R., (2000). “Evaluation of Various Concrete Pavement
Joint Sealants.” Report No. R-1376, Michigan Department of Transportation,
Lansing, MI.

19. Fincher, H.E. (1983). “Evaluation of Rubber Expansion Joints for Bridges.” HPR-1
(18) Part II, Indiana Department of Highways, Indianapolis, IN.

20. Frederick, G.A. (1984). “Development of Water Tight Bridge Deck Joint Seals”.
Report No. 84-7, New York State Department of Transportation, Albany, NY.

21. French, J.W. and McKeel, W.T. (2003). “An evaluation of Bridge Deck Joint Sealing
Systems in Virginia.” Technical Assistance Report VTRC 03-TAR, Charlottesville,
VA.

22. GE Bayer Silicones. http://www.gesilicones.com/gesilicones/eu1/en/home.jsp

(October, 2003), Wilton, CT.

23. Gelest, Inc. <http://www.gelest.com> (October, 2003), Morrisville, PA.

24. Gloyd, S.G., (1996). “Seismic Movement at Bridge Abutments.” The Fourth World
Congress on Joint Sealing and Bearing Systems for Concrete Structures, Sp-164,
AC, pp 273- 287.

25. Guzaltan, F. (1993). Bridge Inspection and Rehabilitation: A Practical Guide, Silano,
L.G., ed., Wiley, NY.

26. Hamilton, C.D. (1985). “Bridge Deck Expansion Joints Final Report.” Report No.
FHWA-ME-TP- 84-04. Maine Department of Transportation, Augusta, ME.

100
27. Hearle, J.W.S. (1982) Polymers and Their Properties, Volume 1: Fundamentals of
Structure and Mechanics, Halsted Press, New York.

28. Hill, J.J. and Shirole, A.M. (1984). “Bridge Joint Systems-A Performance
Evaluation,” in Transportation Research Record 990: Sealing Bridge and Pavement
Joints, Transportation Research Board, National Research Council, Washington D.C.,
pp. 1-4

29. ISO (1991). “Rubber, vulcanized or thermoplastic -- Determination of compression

set at ambient, elevated or low temperatures. ISO 815”, International Organization for
Standardizations, Geneva, Switzerland.

30. James, H. M., Guth, E. (1942). Ind. Eng. Chem. 34, 1365.

31. Kethcam, S.A., (1995). “Sealants and Cold Regions Pavement Seals. A Review.”
Report No. CRREL 95-11, US Army Corps of Engineers, CRREL, Hanover NH.

32. Klafter, J. and Shlesinger, M.F. (1986). "On the Relationship Among Three Theories
of Relaxation in Disordered Systems", Proc. Natl. Acad. Sci. USA 83 848.

33. Kohlrausch, F. W. (1863). “Ueber die elastische Nachwirkung beider Torsion,” Pogg.
Ann. Physik (4) 119, 337-368.

34. Lee, D. J. (1994). Bridge Bearings and Expansion Joints. Alden Press, Osney Mead,
Oxford, Great Britain.

35. Linfante, A., and Castro, A.J. (1977). “Bridge deck Joint Sealing Systems Evaluation
and Performance Specification.” National Cooperative Highway Research
Program Synthesis of Highway Practice 204, Transportation Research Board,
National Research Council, Washington, D.C., 46 pp.

36. Loza, G., and Anderson, D. I. (1987). Evaluation of Concrete Joint Sealants, Clear
Creek Summit to Belknap Interchange. Utah Department of Transportation, Salt Lake
City, Utah.

37. Malla R.B., Shaw M.T., Shrestha, M.R., and Boob, S. (2005a). “Silicone Foam
Sealant for Bridge Expansion Joints.” Proceedings, 2005 Joint ASME/ASCE/SES
Conference on Mechanics and Materials (McMat 2005), Baton Rouge, LA.

38. Malla R.B., Shaw M.T., Shrestha, M.R., and Boob, S. (2005b). “Experimental
Evaluation of Silicone Foam Sealant for Bridge Expansion Joints.” Proceedings,
2005 Society for Experimental Mechanics Annual Conference (SEM 2005), Portland,
OR.
39. Medalia, A.I., and Kraus G. (1994). Science and Technology of Rubber, 2nd Ed.,
(Mark, J.E., Erman, B., and Eirich, F. R., ed.), Academic Press, San Diego, CA.

101
40. Mullins, L. (1969). “Softening of Rubber by Deformation.” J Rubber Chemistry and
Technology, 42(1), 339.

41. Mumber, J. (1993). Bridge Inspection and Rehabilitation: A Practical Guide, Silano,
L.G., ed., Wiley, NY.

42. National Weather Service. http://www.erh.noaa.gov/box/dailystns.shtml (July, 2005),

Taunton, MA.

43. NCHRP (1977). “Bridge Bearings.” National Cooperative Highway Research

Program 41, Transportation Research Board, National Research council, Washington,
D.C.

44. NCHRP (1979). “Bridge Deck Joint-Sealing Systems Evaluation and Performance
Specification.” National Cooperative Highway Research Program 204,
Transportation Research Board, National Research council, Washington, D.C.

45. New Hampshire Department of Transportation (2003). Bridge Maintenance Division,

Concord, NH.

46. Ohio State Department of Transportation (2003). Bridge Maintenance Manual, Ohio
State Department on Transportation, Columbus, OH.

47. Price, A.R. (1984). “The Performance in Service of Bridge Deck Expansion Joints.”
TRRL Report LR 1104, Transportation and Road Research Laboratory, Crowthorne,
U.K., 4-10.

48. Purvis, R. (2003). “Bridge Deck Joint Performance.” NCHRP Synthesis 319,
Transportation Research Board, Washington, D. C.

49. Smith, L.P. (1993). The Language of Rubber: An Introduction to the Specification
and Testing of Elastomers. Butterworth-Heinemann, Oxford, U.K.

50. Stadelmann, B.; Vick, T., (2001), “RTV-2 Silicone Foams: Diverse Material for
Diverse Applications.” International SAMPE Symposium and Exhibition, 46(2001):
A Materials and Processes Odyssey, Book 1, pp 375-379.

51. Stoegbauer, H., and Wolf, A. T. (1990). “The Influence of Heat Aging on One-Part
Construction Silicone Sealants,” Building Sealants: Materials, Properties and
Performance ASTM STP 1069, (O’Connor, T.F., ed.), American Society for Testing
and Materials, Philadelphia, PA.

52. Systat Software Inc. http://www.systat.com, (July, 2004), Richmond, CA.

102
 53. Thirion P., and Chasset R. (1967). “Viscoelastic Relaxation of Rubber Vulcanizates
in Extension.” Chimie & Industrie, Genie Chimique, 97(5), 617-626.

54. Treloar, L.R.G. (1970). Introduction to Polymer Science. WYKEHAM Publications

(London) LTD, London and Winchester.

55. Washington State Department of Transportation (2005). Bridge Design Manual,

Washington State Department of Transportation, Olympia, WA.

56. Watson Bowman Acme Corp. “Wabo®Bridge & Highway Maintenance Products”,
<http://www.wbacorp.com/home_pages/bridge_maintenance.htm> (Nov. 5, 2003),
Amherst, NY.

57. Williams, G. and Watts, D. C., (1970). "Non-symmetrical dielectric relaxation

behavior arising from a simple empirical decay function", Trans. Faraday Soc., 66,
80-85.

58. Vincent, H. L., (1972), “Silicone Foams.” Monographs on Plastics, 1(Pt. 1), pp 385-
396.

59. Voigt, G.F., and Yrjanson, W.A. (1992). “Concrete Joint Sealant Performance
Evaluation.” Final Report, Utah Department of Transportation, Salt Lake City, Utah.

103

View publication stats