Design Guide for Bridges for Service Life (2013)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

543 AppENDix A Design Provisions for Self-Stressing System for Bridge Application with Emphasis on Precast Panel Deck System 544 A.1 Construction Procedure Overview 545 A.2 Design Considerations 546 A.3 Design Procedure and Implementation Details 556 A.4 Design Flowchart 557 A.5 Design Aids for Two-Span Bridges 559 AppENDix B Displacement of Skewed Bridge 559 B.1 Background 563 B.2 Analyses for Transverse Response to Thermal Expansion 566 B.3 Expected Transverse Movement with Typical Integral Abutment 570 AppENDix C Design of Piles for Fatigue and Stability 570 C.1 Estimation of Maximum Allowable Strain 573 C.2 Pushover Analysis Example 576 AppENDix D Restraint Moments 576 D.1 Background 581 D.2 Design Recommendations 587 AppENDix E Design Steps for Seamless Bridge System Developed by SHRP 2 Project R19A 590 E.1 Structural Analysis 590 E.2 Design of System Components 593 E.3 Cracked Section Analysis 596 AppENDix F Curved Girder Bridges 596 F.1 Background 597 F.2 Calculating Magnitude and Direction of End Displacement 600 F.3 Optimum Pile Orientation

602 AppENDix G Design Provision for Sliding Surfaces Used in Bearing Devices for Service Life 602 G.1 Introduction 603 G.2 Elements of Design Provisions 609 G.3 Design Process for Sliding Surfaces

11 DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK The design for service life is gaining more importance as limited resources demand enhancing the service life of existing and new bridges. As part of SHRP 2 Project R19A, Bridges for Service Life Beyond 100 Years: Innovative Systems, Subsystems, and Components, a systematic and general approach to design for service life has been developed. The major product of this project is this document, titled Design Guide for Bridges for Service Life, and referred to here as the Guide. This chapter provides the general framework used in the Guide, which primarily pertains to bridges with spans of less than about 300 ft. However, the framework is general and can be adapted and customized for major and complex bridges, including those with much longer spans. It can also be adapted to suit bridges located in any region in the United States, rec- ognizing, however, that although the framework remains the same for all bridges, the resulting details for service life could be significantly different. 1.1 bAckground Providing safety for the public by having adequate strength for constructed facilities has been the cornerstone of the framework used by engineers for bridge design. This design for strength approach has not been restricted to bridges: it has also been the framework one could find in various building codes. In the case of buildings, however, most structural elements are protected from environmental-type loads, and as a result the strength framework has served this sector of the industry very well. In the case of bridges or pavements, which are constructed facilities exposed to environmental loads, the story is different. Significant changes to contemporary bridge design specifications have also been mainly related to strength issues. The transitions from allowable stress design to load factor design, and more recently to load and resistance factor design (LRFD), reflect

2DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE this line of thinking. It is important to note that in the early 1970s, bridge engineers developed criteria for steel bridge details to protect against fatigue and fracture failure. These were indeed service life design provisions. The strength framework did not prevent visionary engineers such as John Roebling from thinking in terms of service life. A review of bridges that have lasted more than 100 years provides valuable lessons. These bridges are not so much innovative in sys- tem or material, but they have proved to be • Maintainable and well maintained over their 100-year lives as a result of their extreme importance or high capital replacement cost; • Adaptable to changes in functional use, as well as service limit state demands; and/ or • Originally overdesigned. Examples of bridges with long service lives are New York City’s oldest East River bridges, the Brooklyn Bridge (the longest bridge in the world when opened to traffic in 1883), the Williamsburg Bridge (the longest bridge in the world when opened in 1903), and Saint Louis’s Eads Bridge (the first steel bridge, opened in 1874). The Brooklyn Bridge has been well maintained and rehabilitated in a timely manner throughout its lifetime. Initial coatings to protect the bridge’s steel from corrosion did not provide a 100-year life, but cleaning and repainting the bridge did. The metal deck of the Brooklyn Bridge has not survived its 100-plus-year service life, but replacement of the replaceable metal decking has. Figure 1.1 shows the Brooklyn Bridge circa 1890. The Williamsburg Bridge was not as well maintained, as evidenced by its emer- gency closing in 1988. In April of that year, after a thorough inspection revealed cor- rosion of the cables, beams, and steel supports, the Williamsburg Bridge was closed to all vehicular and train traffic for nearly 2 months. After engineers performed emer- gency construction on the bridge and reopened it to traffic, a panel of design experts convened to determine if the Williamsburg Bridge should be replaced or if it should be rehabilitated. In November 1988, after evaluating several alternatives, the New York City Department of Transportation (DOT) determined that the Williamsburg Bridge Figure 1.1. The Brooklyn Bridge from the South Street Seaport, circa 1890. Source: Photo courtesy Library of Congress, Prints and Photo graphs Division, Detroit Publishing Company Collection, LC-D4-90107.

3Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK should be repaired while kept open to traffic. This option was deemed to have the least detrimental impact on motorists and nearby communities. In 1991, the New York City DOT began a major rehabilitation of the Williamsburg Bridge. The program was designed to undo the effects of age, weather, increased traffic volumes, and deferred maintenance. Figure 1.2 shows the Williamsburg Bridge circa 1904. The decision to rehabilitate the Williamsburg Bridge instead of undertaking a costly in-place replacement in downtown Manhattan was made possible by the original con- servative design of the bridge cables. The need to rehabilitate the cables was necessi- tated by a poor corrosion-protection choice. The 1988 inspection of the Williams burg Bridge cables revealed significant corrosion, proving the choice of linseed oil by Leffert L. Buck, the designer of the bridge, to be a relatively poor one. For the Brooklyn Bridge, John Augustus Roebling chose a coating of graphite to protect the individual wires of the bridge cable from corrosion, a choice that provided over 100 years of corrosion protection. Fortunately, the cable design for the Williamsburg Bridge used a factor of safety of resistance divided by load of about five. After the significant loss of section as a result of corrosion was observed in 1988, the factor of safety was deemed adequate, and a cable rehabilitation program to arrest the corrosion was initiated instead of a cable replacement. Thus, the original overdesign allowed the bridge and its cables to continue in service. The Eads Bridge, completed in 1874 and named for its designer and builder, James Buchanan Eads, has proved long lived by being well maintained and readily adaptable. Figure 1.3 shows the Eads Bridge circa 1983. The scale of the bridge was unprece- dented: the more than 500-ft span of the center arch exceeded by some 200 ft any arch built previously. The arch ribs were made of steel, its first extensive use in a bridge. An additional innovation was the cantilever erection of the arches without falsework, the first example of this type of construction for a major bridge. An interesting feature of the history of the Eads Bridge has been its adaptability to varying use. (The Brooklyn and Williamsburg bridges have also seen varied use.) The bridge was originally a railway bridge carrying pedestrians on an upper deck with two rail lines below. It eventually carried vehicular and rail traffic; the last train crossed Figure 1.2. The W illiamsburg Bridge, circa 1904. Source: Photo courtesy Library of Congress, Prints and Photographs Division, Detroit Publishing Company Collection, LC-D4-17414.

4DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE the bridge in 1974. By the early 1990s, traffic on the bridge had dwindled to about 4,000 cars a day, and in 1991, the Eads Bridge was closed. It was briefly unused until 1993, when MetroLink, the region’s new light rail system, began to use the lower deck, which originally served passenger and freight train traffic. In 2003, the upper deck reopened to buses and automobiles. Today, a new lane for pedestrians and bicyclists on the south side of the bridge provides a great place to look at the river and the Saint Louis skyline. The examples of these three 100-plus-year-old bridges illustrates that for bridges to serve a long life, they must be • Resistant to environmental and human-caused hazards; • Maintainable (and subsequently maintained) or relatively maintenance free; and • Adaptable to changes in traveled-way cross section and usage. Traditional approaches for enhancing the service life of bridges used in various codes and specifications, such as American Association of State Highway and Trans- portation Officials (AASHTO) specifications, Eurocodes, or British Standards, are mainly in an indirect form, specifying the use of certain details or properties such as cover thickness, maximum crack width, and concrete compressive strength. Recognizing the importance of design for service life has motivated different agen- cies to undertake new initiatives for developing more formal design approaches for service life, similar to those used for design for strength. However, to date the majority of these efforts have concentrated on addressing concrete durability and service life, and significant advances have been achieved in this field. Designing bridges for service life, however, is more than just addressing service life and durability of concrete. The design for service life for bridges needs to be approached in a systematic, all-inclusive manner rather than as a series of isolated tasks, each independently addressing the service life of a particular portion of a bridge. The interaction among Figure 1.3. The Eads Bridge looking toward Saint Louis and the Gateway Arch, circa 1983. Source: Photo courtesy Library of Congress, Prints and Photographs Division, Detroit Publishing Company Collection.

5Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK strategies for enhancing the service life of different bridge elements, components, and subsystems must be given critical consideration. In addition, a maintenance program, retrofit or replacement options, and management plan should all be part of this sys- tematic service life design approach. In summary, at the design stage the design for service life should be approached as a comprehensive plan capable of providing the owner with a complete picture of what will be necessary for the bridge to achieve its specified service life. The most notable efforts to develop a scientific approach for service life and dura- bility of concrete elements (covering buildings, bridges, and tunnels) were a series of studies carried out between 1996 and 1999 in Europe for the Fédération International du Béton (International Federation for Structural Concrete). One of the products of these efforts was the publication of fib Bulletin 34, Model Code for Service Life Design (FIB 2006). Bulletin 34, however, focused only on concrete service life and durability. Further, caution must be exercised when applying the recommendations of this publi- cation to concrete placed in a horizontal configuration, such as a bridge deck. Although Bulletin 34 has many useful recommendations for designing concrete elements for service life and durability, the application of these recommendations to bridge compo- nents such as bridge decks remains a point of debate (in particular, the use of various solutions to Fick’s second law to predict the rate of chloride ingress through deck con- crete). Bulletin 34 recommendations are believed to be most applicable for concrete in vertical configuration and under compression, such as in sub structure columns or sides of concrete box girders. This same debate can also be extended to the use of some of the available commercial and noncommercial programs that use the fundamental concepts stated in Bulletin 34. Efforts to address service life of bridges are not limited to Europe. A significant number of research studies have been carried out and continue to be performed to develop solutions for various service life issues related to different bridge types. One of the missing elements for designing bridges for service life has been a frame- work that would approach the problem in a systematic manner and provide a com- plete solution in a format that could ensure long-lasting bridges. Individual solutions to problems that historically have reduced service life or to issues involving main- tenance plans, retrofit or replacement plans, bridge management, and life-cycle cost analysis (LCCA) are only components of this systematic framework: they are not the framework itself. The steps within this framework should start at the design stage and should provide the owner with complete information for ensuring the serviceability of the bridge for a specified target service life. It is important for the plan to be transpar- ent and identify the challenges for the period of specified service life at the design stage so that the owner will encounter no surprises. 1.2 objectiveS oF the Guide The main objective of the Guide is to provide information about and define procedures for systematically designing for service life and durability for both new and existing bridges. The cost of addressing service life issues at the design stage is significantly lower than taking maintenance and preservation actions while the bridge is in service.

6DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 1.3 bridge Service LiFe terminoLogy And reLAtionShiPS The following sections define service life–related terms and describe the relationships used in the Guide. 1.3.1 Service Life and Design Life Service life. The time duration during which the bridge element, component, subsys- tem, or system provides the desired level of performance or functionality, with any required level of repair or maintenance. Target design service life. The time duration during which the bridge element, component, subsystem, and system is expected to provide the desired function with a specified level of maintenance established at the design or retrofit stage. Design life. The period of time on which the statistical derivation of transient loads is based: 75 years for the current version of AASHTO LRFD Bridge Design Specifica- tions (2012), referred to throughout the Guide as LRFD specifications. 1.3.2 Bridge Element, Component, Subsystem, and System The term bridge subsystem is introduced by the Guide. The terms bridge element, component, and system are the same as those defined by the Federal Highway Admin- istration (FHWA) National Bridge Inventory. Bridge element. Individual bridge members such as a girder, floor beam, stringer, cap, bearing, expansion joint, railing, and so forth. Combined, these elements form subsystems and components, which then constitute a bridge system. Bridge component. A combination of bridge elements forming one of the three major portions of a bridge that makes up the entire structure. The three major compo- nents of a bridge system are substructure, superstructure, and deck. Bridge subsystem. A combination of two or more bridge elements acting together to serve a common structural purpose, such as a composite girder, which could consist of girder, reinforcement, and concrete. Bridge system. The three major components of the bridge (deck, substructure, and superstructure) combined to form a complete bridge. 1.3.3 Service and Design Life: Basic Relationships Several basic relationships exist between the service lives of bridge components, ele- ments, subsystems and systems, and bridge design life. These relationships are described as follows: • Predicting the service life of bridge systems is accomplished by predicting the ser- vice life of its elements, components, or subsystems. • The design life of a bridge system is a target life in years that is set at the initial design stage and specified by the bridge owner. • The service life of a given bridge element, component, subsystem, or system could be more than the target design service life of the bridge system.

7Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK • The end of service life for a bridge element, component, or subsystem does not necessarily signify the end of bridge system service life as long as the bridge ele- ment, component, or subsystem could be replaced or resume its function with a retrofit. • A given bridge element, component, or subsystem could be replaced or retrofitted, allowing the bridge as a system to continue providing the desired function. • The service life of a bridge element, component, or subsystem ends when it is no longer economical or feasible to repair or retrofit it, and replacement is the only remaining option. • The service life of a bridge system ends when it is not possible to replace or retrofit one or more of its components, elements, or subsystems economically or because of other considerations. • The service life of a bridge system is governed by the service life of its critical ele- ments, components, and subsystems. The critical bridge elements, components, or subsystems are defined as those needed for the bridge as a system to provide its intended function. In general, the service life (ts) of the bridge elements, components, and subsystems should be equal to or greater than the design life (tD) of the bridge system, as defined by Equation 1.1: (ts)C, E, SS ≥ (tD)BS (1.1) where (ts)C, E, SS = service life of bridge component (C), bridge element (E), or bridge subsystem (SS); and (tD)BS = design life of bridge system (BS). The service life of the bridge system is less than or equal to the service life of its governing elements, components, or subsystems, as described by Equation 1.2: (ts)BS ≤ [(ts)C, E, BS]critical (1.2) The service life of the bridge system must exceed or be equal to the target design life of the bridge system, as described by Equation 1.3: (ts)BS ≥ (tD)BS (1.3) 1.4 Guide APProAch to deSign For Service LiFe The Guide approach to design for service life is to provide a body of knowledge re- lating to bridge durability under different exposure conditions and constraints and to establish an array of options capable of enhancing service life. A solution for a

8DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE particular service life issue is highly dependent on many factors that vary from location to location and state to state. A solution also depends on local practices and prefer- ences. Consequently, the Guide is not intended to dictate a unique solution for any specific service life problem or identify the best and only solution. Rather, it equips the reader with a body of knowledge for developing specific solutions best suited to stated conditions and constraints. In applying the Guide framework to a particular bridge, including long-span bridges, an array of solutions can be identified for enhancing the service life of a bridge element, component, or subsystem, and an optimum solution can be identified through the LCCA. The solutions can be based on data collected by local DOTs or agencies responsible for maintaining the bridge and, in order to be complete, the LCCA should include maintenance, retrofit, replacement, and user costs. It is important that the list of assumptions and feasible solutions considered for a particular bridge element, com- ponent, and subsystem be communicated and shared with the owner, especially with respect to the LCCA, so that the entire process is fully transparent. The Guide recognizes that not all bridges can or need to have 100 years of service life. Therefore maintenance, rehabilitation, and replacement are part of the service life design process. The Guide provides the general framework to achieve this objective in a systematic manner that considers the entire bridge system and all project demands. Enhancing the service life of existing and new bridges can be achieved in differ- ent ways. Two examples include using improved, more durable materials and systems during original construction that will require minimal maintenance and improving techniques and optimizing the timing of interventions such as preventive maintenance actions. Interventions can be planned and carried out based on the assessment of indi- vidual bridge conditions and needs, or they can be based on a program of preventive maintenance actions planned for similar elements on a group of bridges. A simple example of a preventive, planned maintenance program might include the following activities: • Washing deicing salts off bridge decks in the spring; • Cleaning debris from bridge-deck expansion joints; • Cleaning debris from bearings and truss joints; • Cleaning drainage outlets; • Spot painting steel structures; • Sealing decks or superstructures in marine environments; and • Sealing substructures on overpasses where deicing salts are used on the roadways below. By acknowledging that service life can be extended by either using more durable, deterioration-resistant materials or by planned intervention, a cost comparison can be made to determine the most cost-effective approach for various environmental expo- sure levels and various levels of available maintenance and preservation actions.

9Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK The following sections provide an overview of the general approach used in the Guide. The discussion is customized for new bridges; however, the Guide approach can be used for existing bridges by eliminating some of the steps used for new bridges. Because the discussions are general and use very simple examples to demonstrate the point of discussion, many intermediate steps are eliminated for the sake of clarity. More detailed procedures and examples are provided in subsequent chapters of the Guide. Three related flowcharts (Figures 1.4, 1.5, and 1.6) are used to demonstrate the general approach used in the Guide. Blocks within each flowchart are numbered and described under the corresponding step number. For example, Step 1 corresponds to Block Number 1 in Figure 1.4. After each flowchart, a brief discussion explains the intent of each block within that flowchart. Customization of the framework introduced in the Guide for a particular bridge could be achieved by developing similar flowcharts, making each step of the process transparent to the owner. For major and complex bridges, various elements of the flowcharts need to reflect specific project requirements. Step 1. The design for service life starts by considering all project demands set by the owner, including the service life requirements, as stated in Figure 1.4. Chapter 2 provides examples of local operational and site requirements, as well as service life considerations needing attention. Step 2. All feasible and preliminary bridge alternatives that satisfy project demands should be developed. For example, one might want to consider steel, concrete, and seg- mental bridge alternatives for a particular bridge. The development of the potential bridge system is carried out in a conventional manner, meeting all the provisions of the LRFD specifications. It is good practice to consider potential service life problems, even at this stage of the design process. It is also feasible to use bridge technologies that do not have a specific design guideline within the LRFD specifications. In such cases, the best available design approach could be used, subject to owner approval. Steps 3 and 4. The next steps in the process consist of evaluating each bridge sys- tem alternative one at a time and considering service life issues related to each element, component, and subsystem of that bridge system. For each bridge element, compo- nent, and subsystem, the Guide provides a framework for incorporating the changes and modifications needed to meet service life requirements. For example, assume that one of the alternative bridge systems to be considered for a particular project is a steel bridge. The designer will first develop the prelimi- nary bridge configurations by using the conventional approaches that meet all LRFD specifications. Using procedures depicted in Blocks 4a, 4b, and 4c, each element, com- ponent, or subsystem of the steel alternative will then be checked against the service life requirements by using the fault tree approach described later in the Guide. These evaluation requirements may lead to changes in the details of the element, component, or subsystem under consideration. For example, the preliminary deck configuration may indicate that 8-in.-thick concrete is sufficient from a strength standpoint. Going through the fault tree corresponding to bridge deck and described in Chapter 4 of the

10 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 2. Identify feasible bridge system alternatives that satisfy design provisions of LRFD specifications. 1. Identify the job and service life requirements. General Steps in Design for Service Life 3. Evaluate all components, elements, and subsystems of the selected bridge system alternatives against service life requirements in the Guide stated in various Guide chapters. 4a. Does specific service life apply? Yes No 4b. Identify mitigation procedure and incorporate changes to bridge configurations. 4c . Are all service life requirements considered? Yes No C Go to A Figure 1.4. General flowchart demonstrating the Guide’s approach for service life design.

11 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK Guide, the designer may change the deck thickness to 9 in. to address potential over- loads, or may specify sealing the bottom of the deck to protect it from salt spray, if the bridge is located along the coastline. For major and complex bridges, most of these fault trees must be customized to meet specific needs and preferred practices. Examples of fault trees and how they work are provided in later sections of this chapter. Steps 5 through 8. At the end of Step 4 and after going through appropriate fault trees for various bridge elements, components, and subsystems, the designer will have developed a bridge system that meets both strength and service life requirements, as illustrated by Step 5 in Figure 1.5. To some extent, changes to configurations of various 6. Determine if the draft configuration of the selected bridge alternative satisfies requirements and that the incorporated changes are compatible. 5. Develop a modified bridge system that meets both LRFD specifications and service life requirements. A 7a. Is configuration okay? No Yes 7b. Make appropriate modifications in components, subsystems, or elements for compatibility. Go to B 8. Develop final configuration of the selected bridge alternative. Figure 1.5. General flowchart demonstrating the Guide’s approach for service life design, starting with A from Figure 1.4.

12 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE bridge elements, components, and subsystems are carried out separately. Thus, there is a need to make sure that these changes are compatible and are not contradictory or overly conservative. Steps 6 and 7 in Figure 1.5 depict this process. For example, in the steel bridge example discussed previously, service life requirements may dictate the use of a jointless, integral abutment system and require metalizing the end of the girder. The designer may then want to consider not metalizing the end of the girder, because leaking joints would be eliminated. Finally, for the selected bridge system alternative under consideration, a final configuration is developed (Step 8) that meets both strength and service life requirements. Figure 1.6. General flowchart demonstrating the Guide’s approach for service life design. starting with B from Figure 1.5. 9d. Identify maintenance plan. 9b. Is service life of the parts greater than service life of the system? Yes No 9c. Identify rehabilitation or replacement plan. 10. Compile all requirements for bridge system alternative and compute life-cycle cost. 9a. Predict service life of various components, subsystems, and elements. B 11a. Are all alternatives considered? No Yes 12. Compare advantages and disadvantages of all final alternatives and select final bridge system. 11b. Consider the next bridge system alternative. Go to C igur eral flowchart demonstrating the Guide’s ppr ach for service life desi n.

13 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK Steps 9 through 12. The next step in the process is to evaluate the service lives of the various bridge elements, components, and subsystems of the bridge alternative under consideration and compare its overall service life with the owner-specified target service design life of the bridge system. For example, the owner may require that the bridge provide 100 years of service life, but the life of a particular bridge element, such as the sliding surface for a bearing, may be limited to 20 years. This situation would require a plan for timely replacement of the sliding surfaces. Regardless, there will be a need for a systematic maintenance plan that could require the designer to identify “hot areas” requiring more detailed inspection and maintenance. Blocks 9a through 9d depict the development of a maintenance plan and/or rehabilitation and replace- ment plan for the bridge system alternative under consideration. The result of this process, as illustrated in Step 10, is a bridge system alternative that meets both strength and service life requirements with an associated maintenance and/or rehabilitation or replacement plan for the bridge. Step 10 also includes an LCCA to consider the final configuration of the select bridge alternative and maintenance plan. The same steps are repeated for all bridge alternative systems, as shown by Step 11. After comparing all alternatives, the designer can recommend which alternative should be used, allowing the owner to make the final selection. As described, the selection of the final bridge system within the framework pro- moted in the Guide is mainly based on service life requirements. Some of the details included in the steps presented, such as fault tree analysis, will be described in later sections of this chapter. A summary of steps for design for service life is provided in Section 1.10 of this chapter. 1.5 orgAnizAtion oF the Guide Included in the Guide are 11 chapters, each devoted to particular bridge elements, components, subsystems, or systems. The following is a brief description of informa- tion included in each chapter. Chapter 1. Design for Service Life: General Framework. This chapter provides an overview of the approach used in the Guide for design for service life and describes ter- minology used throughout the Guide and various relationships that exist between the service life of bridge element, component, subsystem, and system and bridge design life as used in AASHTO specifications. The chapter introduces the different philosophies used to predict service life. Chapter 2. Bridge System Selection. This chapter describes various bridge systems and factors that affect their service life. Included is a description of a general strategy and rational procedure for selecting the optimum bridge system, subsystems, compo- nents, and elements that consider specific project limitations and requirements, such as climate, traffic, usage, and importance. Chapter 3. Materials. This chapter provides general properties and durability characteristics of the two most commonly used materials in bridge systems, steel and concrete. For each material, a general description of variables affecting the service life is provided, followed by strategies used to mitigate them. Chapter 3 comprises the

14 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE basic information for materials used in bridge subsystems and elements specifically addressed in other chapters of the Guide. Chapter 4. Bridge Decks. This chapter describes various bridge-deck types and essential information related to their service life, such as modes of deterioration and strategies to mitigate them. The chapter concentrates on cast-in-place and precast con- crete bridge decks. Chapter 5. Corrosion of Steel in Reinforced Concrete Bridges. This chapter looks at basic mechanisms causing corrosion of reinforcement embedded in concrete and provides strategies for preventing corrosion of reinforcement in concrete bridges. Chapter 6. Corrosion Prevention of Steel Bridges. This chapter describes various coating systems using paint, galvanizing and metalizing, and corrosion-resistant steel, along with factors affecting service life. Various options for preventing corrosion of steel bridges and general approaches that could lead to bridge coatings with enhanced service life are presented. Chapter 7. Fatigue and Fracture of Steel Structures. This chapter provides the basics of fatigue and fracture and the factors that cause fatigue and fracture in steel bridges. Various available options for repairing observed cracking in steel bridges are also presented. Chapter 8. Jointless Bridges. This chapter describes various jointless bridge sys- tems, considers their advantages and disadvantages, and provides complete steps for the design of jointless integral abutment bridges. Design procedures to extend the application of jointless integral bridges to curved girder bridges are provided. Also introduced are new details and integral abutment systems, in which expansion joints are completely eliminated, even at the end of approach slabs. Chapter 9. Expansion Devices. The Guide encourages eliminating the use of expansion joints. However, expansion joints may be needed when the total bridge length exceeds the practical limits of jointless bridges. This chapter describes various expansion joints used in practice, observed modes of failure for each, and potential strategies to mitigate those failures. Chapter 10. Bridge Bearings. This chapter describes various bearing types, lists the factors that affect the service life of the various bearings, and offers strategies to mitigate such factors. New high-performing sliding surfaces capable of providing long service life are introduced, as well as deterioration models for sliding surfaces. The Guide emphasizes use of elastomeric bearing pads. Chapter 11. Life-Cycle Cost Analysis. This chapter provides essential information for incorporating LCCA into bridge system, subsystem, component, and element selec- tion. It concentrates on general features and elements of incorporating LCCA into the design process, emphasizing consideration of project costs throughout the service life. 1.6 cAtegorieS oF inFormAtion Provided in Guide chAPterS Typically, each chapter consists of the five major categories of information described in this section. Closer examination of the type of data included in each chapter could also assist in developing customized information for addressing design for service life for major and complex bridges.

15 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK 1.6.1 Description of Bridge Elements, Components, and Systems These sections of each chapter provide brief descriptions of, and essential information related to, both commonly used and more recently developed types of bridge compo- nents, elements, subsystems, and systems. 1.6.2 factors that Affect Service Life The factors affecting service life are identified using a fault tree approach, which pro- vides a systematic method of identifying factors that can affect the service life of a particular bridge element, component, or subsystem in various categories and suc- cessive subcategories. Most chapters have fault trees applicable for the types of ele- ments, components, or subsystems covered within that chapter. In the case of major and complex bridges, designers should develop customized fault trees that reflect the specifics associated with location and traffic conditions. A customized fault tree can be developed using data and experiences available from local agencies. The fault tree starts with the identification of major factors that can reduce the service life of a particular bridge element, component, or subsystem. Each major factor can then be broken down into more detailed subcomponents, each capable of reducing the service life. The fault tree continues branching until each branch ends with factors at the lowest or base levels of influence. The factors with subcomponents are placed inside rectangles, and the identified lowest or base factors are placed inside circles. Figure 1.7 shows a portion of the fault tree used in Chapter 4 for a bridge deck. In Figure 1.7, either of two main factors (obsolescence or deficiency) is shown to be capable of contributing to reduced service life of a bridge deck. The elliptical symbol just above these two factors is referred to as an “or gate,” which signifies that either one of the factors below it could result in reduced service life. The fault tree shown in Figure 1.7 continues to list the major categories of factors (i.e., those related to induced loads, natural or man-made hazards, and production or operation defects) that could result in reduced service life of a bridge deck. Figure 1.8 shows the continuation of the fault tree for a breakdown of factors related to load-induced factors for a bridge deck. In Figure 1.8, load-induced factors are subcategorized into traffic-induced loads or loads induced by system-dependent load factors, such as restraints provided by shear studs. Two factors are further divided into subcomponents, each capable of reducing the service life of a bridge deck. The factors inside the circles are the basic factors without any further subcomponent. They represent the end of that branch of the fault tree and require the development of individual strategies to mitigate them. This aspect of the process is described later. In the Guide, each element of the fault tree is described immediately after intro- ducing each branch of the fault tree. It is advisable to do the same when develop- ing customized fault trees for major and complex bridges. Documentation of factors affecting the service life of bridge elements, components, or subsystems in the form of a fault tree should be part of the overall plan for design for service life and provided to the owner for future reference.

16 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 1.6.3 mitigation Strategies When possible, each chapter provides provable solutions for major factors affecting the service life of a particular bridge element, component, subsystem, or system. Some chapters also include technology tables that summarize major characteristics associ- ated with each solution and provide the potential solutions to factors affecting service life in a form that is easier to comprehend. For example, Figure 1.9 shows a technol- ogy table that summarizes solutions for enhancing the service life of bridge decks; it is related to traffic-induced loads as shown in Figure 1.8. In Figure 1.8, as part of the fault tree for a bridge deck, traffic-induced loads are identified as one factor capable of reducing service life. In Figure 1.8, below traffic-induced loads, three basic factors capable of reducing bridge-deck service life are identified: fatigue, overload, and wear and abrasion. Each basic factor needs to be mitigated using a select strategy, and in almost all cases, there is more than one strategy to mitigate these basic factors. It is good practice to collect these strategies in table form and select the optimal strat- egy, considering its interaction with other parts of the bridge. The technology tables Reduced Service Life of Cast-in-Place Bridge Deck Caused by Deficiency Caused by Obsolescence Natural or Man-Made Hazards Load- Induced Production/ Operation Defects Figure 1.7. Starting point for fault tree for a bridge deck.

17 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK provided in various chapters of the Guide summarize strategies that can be used to mitigate various basic factors capable of reducing service life. For major and complex bridges, the list of strategies could be different and based on local preferences and experi ences. Most agencies have access to field data collected over the years that could be used to construct customized strategy tables for the purpose of mitigating basic fac- tors capable of reducing service life of bridge elements, components, and subsystems. Figure 1.8 shows the continuation of the fault tree for a breakdown of factors related to load-induced factors for a bridge deck. Load-Induced WearFatigue System- Dependent Loads Differential Shrinkage System- Framing Restraint Traffic- Induced Loads Thermal Overload Figure 1.8. Continuation of the fault tree for a breakdown of factors related to load-induced factors for a bridge deck. Service Life issue mitigating Strategy Advantage Disadvantage Fatigue Design per LRFD specifications Minimizes the possibility of reinforcement failure May increase the area of steel Overload Increase deck thickness Minimizes cracking Adds weight to bridge structure, increases cost Wear and abrasion Implement concrete mix design strategies See Chapter 3, Materials See Chapter 3, Materials Implement membranes and overlays Protects surface from direct contact with tires Requires rehabilitation every 10 to 20 years Figure 1.9. Technology table for mitigating factors related to traffic-induced loads affecting the service life of bridge decks.

18 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE The sample table in Figure 1.9 lists the advantages and disadvantages for each possible solution capable of mitigating the adverse service life consequences of traffic- induced loads. In some cases, more information than just advantages and disadvan- tages is provided, such as qualitative assessment of maintenance cost. For major and complex bridges additional considerations may be included in technology tables. 1.6.4 optimum Selection Strategies Overall strategies are provided for achieving enhanced service life. The overall strategy approach provided depends on the particular bridge component, element, subsystem, or system. Figure 1.10 shows an example taken from Chapter 10 of the Guide of the overall strategy for selecting a bearing that meets both strength and service life requirements. Chapter 10, Bridge Bearings, identifies factors affecting the service life of bearings and provides potential solutions for each. This information, combined with the steps outlined in the flowchart, can be used as a rational approach for selecting an appropri- ate bearing that meets project requirements with emphasis on service life. 1.6.5 Examples and tools Most chapters include examples demonstrating the application of strategies in that chapter. 1.7 QuAntiFying Service LiFe oF bridge eLement, comPonent, SubSyStem, And SyStem One of the important steps in developing a systematic, comprehensive service life d esign plan for bridges is the capability of predicting the expected service life of various bridge elements, components, and subsystems, which in turn will dictate the service life of the bridge system. This process is Step 9a in Figure 1.6. Service life prediction capability is important for developing maintenance, retrofit, and replacement plans, which are an integral part of the service life design process. The objective of this section is to provide an overview of the methodology used in the Guide for predicting service life. Bridge elements, components, subsystems, and systems are subject to the effects of traffic and the environment. These external sources of deterioration act through vari- ous mechanisms to cause actual deterioration and eventually failure of bridge elements. The mechanisms of deterioration are the physical laws that govern such deterioration. Deterioration rates can be described using mathematical expressions or empirical and semiempirical models, which are developed using data collected by field monitoring of bridges, laboratory-generated data, expert opinions, or combinations of available data. Service life is also affected by risk to damage from either traffic or extreme envi- ronmental occurrences. The acceptability of this damage is evaluated based on risk. Service life can be extended by minimizing risk or designing for appropriate levels of extreme occurrences.

19 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK Enhanced service life for bridge elements, components, subsystems, and systems can be achieved through • Use of durable materials; • Use of passive or active protection systems; Identify DEMAND requirements. Compare with SUPPLY parameters for bearing alternatives. Identify potential bearing alternative(s) that accommodate loads and movements. Perform preliminary design to confirm suitability of potential bearing type. No Perform final bearing selection. Yes Determine expected frequency of bearing replacement. Identify factors affecting service life and mitigation requirements. Compare life-cycle costs considering initial, maintenance, and replacement costs. Is bearing suitable? Evaluate potential service life of alternatives and determine if bearing has potential for achieving desired life. No Yes Is bearing suitable? Figure 1.10. Overall strategy for bearing design considering service life.

20 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE • Optimum selection of details; • Optimum maintenance and repair; • Reduced service level; • Increased factor of safety or reduction in stress levels; and • Isolation from risk damage. To estimate the service life of bridge elements, components, or subsystems quanti- tatively, the following information is needed: • Source of deterioration; • Deterioration mechanism; • Deterioration models; and • Failure modes. The following sections provide information on each of these items of information. 1.7.1 Source of Deterioration Traffic-related or environmental effects are the basic external causes of deterioration. For example, deicing compounds, an external source of deterioration, can result in corrosion of reinforcement in bridge elements. 1.7.2 Deterioration mechanism Deterioration is governed by a process called the deterioration mechanism. For ex- ample, sliding surfaces in bearings experience deterioration through horizontal move- ment and friction between sliding materials created by truck passages or temperature fluctuations. The horizontal movement and friction in this instance is the deterioration mechanism. In the case of concrete elements, ingress of chloride through concrete causes initiation of corrosion in unprotected steel reinforcement. In this instance, the ingress of chloride is the deterioration mechanism. 1.7.3 Deterioration models Deterioration models are used to describe the rate of deterioration. They describe the relationship between the condition of the bridge (or its element) and its time of use and show how the bridge deteriorates over time. A deterioration model assumes that no replacements or major repairs are made, but it usually implies that scheduled main- tenance actions are performed as planned. The basic model applies either to a bridge system as a whole or to any of its subsystems, components, or elements. An example of a deterioration curve is presented in Figure 1.11. For a bridge placed in service at period T0, the deterioration curve represents the gradually declin- ing condition of the bridge over time. Initially the condition is good, but after a period of wear and aging, it eventually (at time Tf) reaches an unacceptably low condition (Cf). The time period between T0 and Tf is called the service life of the bridge.

21 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK In practice, the development of realistic behavioral deterioration models is a data- intensive process complicated by lack of knowledge of the underlying physical and chemical processes fostering deterioration, as well as by data availability. Currently available deterioration models, which are based on long-term data collection, are very limited. To complicate the issue, the quality of bridge design and construction improves over time. As a result, application of data collected from existing bridges to predict the performance of future bridges should be practiced with caution. Deterioration models capable of quantitatively predicting the service life of bridge elements, components, subsystems, or systems are very limited or nonexistent. The most acceptable deterioration model is in the form of the solution to Fick’s second law, which is used to predict the rate of chloride ingress through concrete cover. This model, including its limitations, is described in Chapter 5 of the Guide. It is expected that with time, more deterioration models will become available and will greatly enhance quan- tification of the service life of bridge elements, components, subsystems, or systems. As shown on Figure 1.12, if left alone a bridge will deteriorate over the period of its service life. However, in most cases a bridge is not left to follow the basic deterio- ration path and reach an unacceptable condition without interruption. The agency responsible for the bridge will from time to time undertake repairs, rehabilitations, and renewals that return conditions to higher levels and extend its service life. During these interventions, the condition of the bridge improves, as depicted in Figure 1.12. Deterioration models can also be based on some level of understanding of the mechanism governing the deterioration and the capability of expressing the process through a mathematical expression. An example is deterioration of concrete elements caused by chloride-induced corrosion of reinforcement. The assumption is that ingress of chloride through the concrete element is governed by Fick’s second law, which assumes a homogeneous material. Bridge Deterioration Service Life T 0 Cf Tf Condition Time of Use Figure 1.11. An example of a bridge deterioration curve.

22 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE In the case of chloride- and carbonation-induced corrosion, there is some level of agreement within the scientific community as to the existence of deterioration models. However, for other deterioration modes, such as sulfate attack, alkali-silica reactiv- ity (ASR), and freeze–thaw or wear and abrasion, there is a lack of adequate models. Further, as described previously, the use of deterioration models to predict the time to initiate corrosion of reinforcement embedded within certain distances of the concrete surface because of chloride ingress should be approached with caution. The following paragraphs briefly describe a deterioration model for chloride-induced corrosion. There are different approaches to solving Fick’s second law. A finite-difference approach, or the use of error functions, is reported in the published literature. Equa- tion 1.4 is an error function solution of Fick’s second law that is capable of predicting the chloride concentration level at various depths within the concrete element. ( )( )= = + − − − ∆         ∆C C x t C C C a x D t , . 1 erf 2.S x C crit 0 , 0 app, (1.4) where Ccrit = critical chloride content (wt.-%/c), C(x,t) = content of chlorides (wt.-%/c) in the concrete at depth x (structure surface: x = 0 m) and time t, C0 = initial chloride content (wt.-%/c) of the concrete, CS,Dx = chloride content (wt.-%/c) at depth Dx and certain point of time t, x = depth (mm) with a corresponding content of chlorides [C(x,t)], a = concrete cover (mm), Dx = depth of convection zone (concrete layer, up to which the process of chloride penetration differs from Fick’s second law of diffusion) (mm), Dapp,C = apparent coefficient of chloride diffusion through concrete (mm 2/years), t = time (years), and erf = error function. Bridge Condition Service Life T 0 Cf Time of use Condition Tf Figure 1.12. Bridge condition life cycle.

23 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK Equation 1.4 should be used in conjunction with probabilistic approaches to account for the variability of several parameters, such as apparent coefficient of dif- fusion, chloride concentration, and critical chloride level to start corrosion. Further- more, the diffusivity of concrete through different layers of the concrete element is not uniform. Equation 1.4 predicts the chloride content in the structure at a given depth x and time t. This number is given by the left-hand side of the equation, C(x,t). The C(x,t) obtained from Equation 1.4 is then compared with the critical chloride content (Ccrit), which is the value determined to be the point at which corrosion starts. When the chloride level at a given depth (x) of the structure is reached (i.e., the critical value), the corrosion is assumed to initiate. The service life of the concrete element can then be assumed to consist of the time period to initiate corrosion plus the time period for propagation of the corrosion to the point that will limit the functionality of the concrete element. This process is depicted in Figure 1.13. 1.7.4 failure modes Sources of deterioration (such as deicing compounds) acting through deterioration mechanisms (such as ingress of chloride through concrete cover) and described by deterioration models (such as a solution to Fick’s second law) result in failure modes (corrosion of reinforcement, causing corrosion-induced cracking and loss of strength). The final failure could consist of several stages, such as start and propagation phases. 1.7.5 Service Life Estimation In the Guide, two general philosophies are presented to estimate the service life of bridge elements, components, and subsystems. Quantifying the service life of bridge ele- ments, components, or subsystems establishes ts, the design service life of bridge ele- ments, components, or subsystems, which can then be compared with the specified service life of the bridge system to determine whether retrofit or replacement strategies are needed. Figure 1.13. Relationship between damage and service life. Source: Edvardsen 2008.

24 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Two general design approaches for service life are the finite service life approach and the target service life approach. When the design service life (ts) of the bridge element, component, or subsystem established through one of the two design approaches for service life philosophies is less than the specified design life of the bridge system (tD), the bridge element, com- ponent, or subsystem under consideration could be replaced to achieve the specified design life of the bridge system. The major difference between the two approaches for service life design is the need for well-accepted deterioration models in the finite service life approach. 1.7.5.1 Finite Service Life Approach Bridge elements, components, and subsystems designed using the finite service life ap- proach should have a service life that is greater than or equal to the specified bridge system service life. Otherwise, the bridge element, component, or subsystem under consideration must be retrofitted or replaced to allow the bridge to continue provid- ing its intended function until reaching the specified bridge system service life. In the finite service life approach, the service life of the bridge components, elements, or sub systems is estimated using well-accepted deterioration models. The existence of deterioration models is therefore essential for using the finite service life approach. The deterioration models are generally developed using one of the following approaches: • Mathematical models that describe the deterioration rate. These models could be approximate or based on the laws of physics; • Empirical or semiempirical models developed using data collected from laboratory or field performance of bridges. Fatigue models used in the LRFD specifications are examples of empirical deterioration models; or • Empirical models based on expert opinions or experiences. Examples include vari- ous models used in Pontis. When deterioration models exist, the service life design could be in the form of a full probabilistic approach or a semiprobabilistic or partial load factor approach. The full probabilistic approach requires having probability distribution functions for all variables used in the deterioration model. The semiprobabilistic or partial load factor approach is developed using the full probabilistic approach. It is equivalent to using load and resistance factors in the LRFD specifications versus using the full probabilistic approach (such as using Monte Carlo simulation) to design or rate bridges. 1.7.5.2 Target Service Life Approach In many instances deterioration models are not available, or their applicability is ques- tionable. In these situations, available alternatives are (1) the use of high- performing material that does not deteriorate, such as stainless steel (this approach is generally referred to as the avoidance of deterioration method within European practice) or (2) the use of material that, based on experience or expert opinion, could provide

25 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK a specified or target service life. If the estimated service life of the bridge element, component, or subsystem is less than the specified service life of the bridge, retrofit or replacement strategies must be specified, allowing the bridge system to continue providing its intended function. The major difference between the finite service life and target service life design approaches is that in the finite service life design approach, the condition of the bridge element, component, and subsystem can be traced over time by using deterioration models; in contrast, in the target service life design approach, only the total expected service life is estimated. The specified target service life of the bridge element, compo- nent, or subsystem is mainly established on the basis of experience or expert opinion and could vary significantly from assumed values. Nevertheless, specifying a target ser- vice life for a given bridge element, component, or subsystem allows the bridge owner to plan and anticipate necessary maintenance actions and places demands on the designer to incorporate necessary design features as needed. For example, in response to an assumed service life of about 10 years (i.e., target service life of 10 years) for polytetrafluorethylene sliding surfaces in bearing devices, the designer must incorpo- rate necessary mechanisms to lift the bridge and replace the sliding surfaces, prefer- ably while maintaining traffic, and the bridge owner must plan for and anticipate the replacement of sliding surfaces every 10 years. 1.8 owner’S mAnuAL When specified by the owner and for major and complex bridges, the final step in the design for service life process is the development of a bridge Owner’s Manual, which summarizes the processes used for the design for service life and provides complete descriptions of outcomes and recommendations. The Owner’s Manual is intended to equip the owner with the necessary knowledge to keep the bridge operational for the specified service life period. It should be provided to the owner at the time of opening the bridge to traffic, after an independent review process as described in Section 1.9. The entire process used for design for service life should be well documented and include assumptions, limitations, and any other information about which the owner should be aware, including complete information with respect to “hot spots” within various bridge elements, components, or subsystems that will require closer inspection, maintenance, retrofit, or replacement. The Owner’s Manual should include a complete management plan with respect to service life, including information on timely mainte- nance actions, and identify replacement items and methodologies for replacement with information on the required level of traffic interruption, if any. In the case of major and complex bridges it is suggested that a bridge instrumentation and monitoring plan be developed and tied to the bridge service life management plan. Additional informa- tion to be incorporated into the Owner’s Manual after construction should include the actual material properties of critical bridge elements versus the assumed values used in the design process. Such information is important for future bridge rating. For major and complex bridges, the designer should use sound engineering judg- ment for determining the level and extent of information to be included in the Owner’s Manual. The bridge Owner’s Manual is analogous to the design calculations that are

26 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE customarily provided to the bridge owner, except that the Owner’s Manual contains much more detailed information. 1.9 indePendent review oF deSign For Service LiFe ProceSS The design for service life processes, results, and recommendations as summarized in the bridge Owner’s Manual should be checked by an independent and knowledgeable third party. This independent check is analogous to an independent design check typi- cally conducted for bridge design. 1.10 SummAry oF StePS For deSign For Service LiFe For SPeciFic bridge eLement, comPonent, And SubSyStem This section summarizes the steps in the design for service life. Detailed descriptions of individual steps are provided in Section 1.4. Bridge elements, components, and subsystems can deteriorate at different rates and have different service lives. This variable deterioration governs the service life of a bridge system, which is reached when the service life of critical bridge elements, com- ponents, or subsystems is exhausted beyond being repaired or replaced economically or because of other considerations. The general steps in design for service life for a particular bridge element, compo- nent, or subsystem can be summarized as follows: Step 1. Identify the project requirements, particularly those that will influence the service life. Step 2. Identify feasible bridge systems capable of meeting the project demand. Step 3. Select each feasible bridge system one at a time and complete Steps 4 through 10. Step 4. Identify the factors that influence the service life of bridge elements, com- ponents, and subsystems, such as traffic and environmental factors. Step 5. Identify modes of failure and consequences (e.g., the corrosion of rein- forcement that causes corrosion-induced cracking and loss of strength). Step 6. Identify suitable approaches for mitigating the failure modes or assessing risk of damage through LCCA (e.g., use of better-performing materials for sliding sur- faces in bearings or use of material prone to deterioration at lower initial cost). Step 7. Modify the bridge element, component, or subsystem under consideration by using the selected strategy and ensure compatibility of different strategies used for various bridge elements, components, or subsystems. This step may involve the need to develop several alternatives. Step 8. For each modified alternative, estimate the service life of the bridge ele- ment, component, or subsystem using the finite or target service life design approaches. Step 9. For each modified alternative, compare the service life of the bridge ele- ment, component, or subsystem with the service life of the bridge system and develop appropriate maintenance, retrofit, and/or replacement plans.

27 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK Step 10. For each modified alternative, develop design, fabrication, construction, operation, maintenance, replacement, and management plans for achieving the speci- fied design life for the bridge system. Step 11. For each modified alternative, conduct LCCA for each feasible bridge system meeting strength and service life requirements and select the optimum bridge system. Step 12. When specified by the owner or in cases of major and complex bridges, document the entire design for service life processes in an Owner’s Manual. Arrange for an independent review of the document and provide it to the bridge owner at the time of opening the bridge to traffic. 1.11 APProAcheS to uSing the Guide This section provides a limited example demonstrating the use of the Guide and how to implement systematic approaches for design for service life. The example is not inclusive and considers an isolated component of the bridge without considering the remaining bridge elements, components, or subsystems. Further, the example, for the sake of demonstration, uses Life-365, which has limitations when applied to horizon- tal surfaces, such as bridge decks. Life-365 uses the solution to Fick’s second law to predict deterioration of concrete elements subjected to chloride ingress. Although this approach has merits for vertical surfaces, such as columns under compression, its ap- plicability to horizontal surfaces, such as bridge decks, is not warranted. In the case of bridge decks, the existence of cracks violates the assumption of a homogeneous mate- rial in Fick’s second law. The use of Life-365 for the bridge-deck example here is for demonstration purposes, as it includes LCCA in addition to predicting time to initiate corrosion and propagation. The following sections provide an overall description of the bridge used for the example and illustrate the steps taken in the design for service life for a single isolated component of the bridge. 1.11.1 Example Bridge Description The example bridge is a 1,400-ft-long structure carrying four lanes of high-volume traffic with pedestrian sidewalks and bicycle lanes. The bridge crosses over low- volume urban local roads, a railroad, and a navigable waterway. Refer to Figures 1.14 and 1.15 for a rendering of the project concept. Figure 1.16 shows the bridge-deck system that will be used for this example. The following characteristics of the bridge setting influence the service life: • Located in a cold environment where deicing salts are used and multiple freeze– thaw cycles are anticipated; • Located in an area where studded tires are used in the winter; • Subjected to potential overloads with 20-kip tire loads in an HL 93 truck configuration;

28 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Figure 1.14. Aerial conception of bridge project. Figure 1.15. Typical superstructure and substructure configuration.

29 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK • Spans over a navigable waterway with primarily brackish conditions, and located adjacent to a park with water access for jet skis; • Located near the coastline with possible salt water storm surge and potential hur- ricane force winds with gusts up to 150 mph; and • Located in an area where local aggregates are subject to ASR. 1.11.2 Steps in Addressing Service Life Design The first step in the process for addressing the service life design issue for a bridge deck is to follow the flowchart shown in Figure 1.17; this flowchart also appears as Figure 4.18 in Chapter 4 of the Guide. Table 1.1 is used to extract the information needed to address the requirements of Steps 1a and 1b shown in Figure 1.17. The information in Table 1.1 was developed from project requirements and exem- plifies the type of information necessary for layout and service life evaluation of the entire bridge system. It is well beyond that information needed for simply evaluating a bridge deck, but it is provided here for completeness. The issues in Table 1.1 pertinent to the bridge deck are indicated by an arrow at the right side of the table. The next step is to identify the possible bridge-deck alternatives (Step 2 in Fig- ure 1.17). Information provided in Section 2.3.2 of Chapter 2 and Section 4.2.1 of Chapter 4 of the Guide can be used to obtain advantages and disadvantages of vari- ous deck systems. In the case of major and complex bridges the designer may consider feasible bridge-deck systems on the basis of local preferences and experiences. In the Figure 1.16. Superstructure section of bridge deck with cast-in-place option designed according to LRFD specifications.

30 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 1b. Identify local factors affecting service life. 2. Identify feasible deck alternatives satisfying design provisions of LRFD specifications, operational, site, and bridge system requirements. 1a. Identify local operational and site requirements. 3. For each alternative, identify factors affecting service life by following fault tree. Go to A Bridge-Deck System Component Selection Process Yes No 6. Identify maintenance requirements. 5a. Identify rehabilitation or replacement requirements. Yes 7. Develop life-cycle costs. No 5. Is deck service life greater than or equal to the system TDSL? 8. Additional deck alternative? 9. Compare alternatives and select deck system. B 8a. Go to the next alternative. 4. Figure 1.17. Bridge-deck system component selection process. For steps from A to B, see Figure 1.18. (TDSL means target design service life.)

31 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK tABLE 1.1. oPerAtionAL And LocAL FActorS to be conSidered: StePS 1a And 1b in Figure 1.17 Operational Category Operational Criteria to Be Specified Traffic capacity requirements Urban arterial, four lanes, 40 mph Traffic volumes and required capacity 24,000 average daily traffic northbound and southbound á Truck volumes 10% á Special vehicle uses Overload possible á Local environment or man-made hazard category Maintain two existing lanes á Mixed-use requirements Traffic, pedestrians, bicycle lane Vehicle loads and special vehicle load requirements HL 93 with typical legal and permit loads No special construction loads Overload with 20-kip tire loads (HL 93 truck configuration) Studded tires used in winter á Service Life Category Service Life Criteria to Be Specified Identify bridge importance Critical Identify target design service life 100 years á Potential for future bridge widening Not applicable Potential for future widening of crossed roadways Not applicable Vertical clearance requirements related to future bridge widening or widening of crossed roadways Not applicable Local Site Category Local Site Criteria to Be Specified Geometry See plan. Overall bridge length: 1,400 ft Curvature: Curve 1 R = 1,150 ft; Curve 2 R = 1,300 ft, reversing Cross slope: superelevation transition from +2% to –2% Skew: 35° Features crossed Road A: two lane urban, 25 mph Road B: one lane urban, 25 mph (two lanes temporary) Railroad: one existing track, four additional future tracks, potential commuter rail corridor with mixed freight usage Navigable waterway: U.S. Coast Guard jurisdiction, small vessels and jet skis Horizontal clearances Standard clear zones with no barrier protection No railroad crash walls allowed No piers in waterway (continued)

32 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Local Site Category (continued) Local Site Criteria to Be Specified (continued) Vertical clearances Road A: 16.5 ft plus 2 ft for raising of tracks (for railroad bridge replacement) Road B: 16.5 ft Railroad: 23.5 ft plus 2 ft for raising of tracks (for railroad bridge replacement) Navigable waterway: 15 ft minimum (bank to bank) Hydraulic or waterway requirements Greatest flood (500-year) elevation 7.50 (storm surge) Natural bend in channel Historic erosion of south bank Navigation requirements No piers in waterway (bank to bank) Utility issues, carried or crossed Bridge lighting Relocate underground fiber optic lines Other physical boundary conditions Maintain access to adjacent park Geotechnical considerations All foundation types acceptable Environmental considerations Water in waterway tests with low chlorides, but adjacent mangroves indicate that water may be brackish at times Subject to salt water intrusion from storm surge Access for construction Limited to existing right-of-way and railroad agreement Aesthetics and sustainability Closed box system required by city (I-girders unacceptable) Local Environmental or Man-Made Hazard Category Local Environmental or Man-Made Hazard Criteria to Be Specified Thermal climate Cold climate, solar radiation, Zone 3 Deicing salts used, multiple freeze–thaw cycles, ice flow Coastal climate Brackish conditions Chemical climate ASR susceptible Hydraulic action hazard, flood, or scour 500-year water velocity/discharge: 4.8 fps/11,200 cfs Wind action hazard Hurricane zone (150 mph), Exposure C Drainage requirements 50-year storm Vehicle and vessel collision susceptibility Vehicle collision to be addressed Vessel collision potential negligible (mostly pleasure craft) Fire or blast susceptibility Minimal combustible materials on route Seismic susceptibility Seismic Zone 1 Construction Constraints Category Construction Constraints Criteria to Be Specified Constructability requirements Phasing not required Construction schedule requirements Accommodate short schedule demands Special local construction preferences Do not use cast-in-place concrete boxes Note: Rows identified by á point out items used in developing the design example listed in this section. tABLE 1.1. oPerAtionAL And LocAL FActorS to be conSidered: StePS 1a And 1b in Figure 1.17 (­continued)

33 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK current example, it is assumed that only the cast-in-place option is selected, as indi- cated in the typical girder cross section shown in Figure 1.16. Figure 1.18 shows the next steps in identifying factors that affect service life; the same flowchart appears as Figure 4.19 in Chapter 4 of the Guide. Figure 1.18 aids in identifying factors that affect the service life of the bridge deck and in selecting possible strategies capable of mitigating the adverse effects of these factors. Identifying the factors that affect bridge-deck service life can be accom- plished using the fault trees provided in Chapter 4, Section 4.2 of the Guide. Navigat- ing through the fault tree can be simplified by using software. Figure 1.19 shows an example of what an Excel-based solution could look like. Using the software shown in Figure 1.19, the user selects applicable factors from the first fault tree layer and then continues through successive layers until reaching the last levels, which are depicted as circles. The items listed in each circle are the factors that will have to be addressed in Figure 1.18. Flowchart to identify factors affecting service life (Figure 4.19). 2A.b. Modify bridge deck configuration. A 2A.a. Identify consequences and determine appropriate strategies for avoidance or mitigation. 1A. Identify individual factors affecting service life considering each branch of fault tree. 3A.a. Go to the next factor. 4A.Modified bridge deck configuration for deck alternative under consideration. Go To B Yes No 2A. Does factor apply? Yes No 3A. Are all factors considered? 4. Figure 1.18. Flowchart to identify factors affecting service life.

34 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Figure 1.19. Screenshot of navigation through fault tree using an Excel worksheet. the design for service life process. Each factor has the capability of reducing the service life of the bridge deck. Chapter 4 identifies one or more strategies capable of mitigating the effect of each particular factor listed in the circles. Figure 1.19 shows branches of the fault tree that are applicable to the example under consideration. In the first layer, based on the project requirements, a decision is made that the service life of a bridge deck can only be reduced due to deficiencies. The second layer states that either loads or natural or man-made hazards can cause bridge- deck deficiencies. Both are judged to be applicable and are therefore selected. Figure 1.19 then shows the progression through the fault tree on the branch related to the fac- tor “Due to Loads.” The fourth layer in Figure 1.19 states that either traffic-induced

35 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK loads or system-dependent loads can reduce bridge-deck service life. Again, both fac- tors are judged to be applicable, and the boxes are checked. Further, the reduction in service life of the bridge deck as a result of traffic-induced loads can be caused by fatigue, overload, or wear, as depicted by the circles. All circles are applicable to the example. Figure 1.19 also identifies each factor that is capable of reducing the service life of the bridge deck as a result of system-dependent loads, as shown in circles and identified as differential shrinkage, thermal, and system-framing restraint. Based on project requirements, these factors are also judged applicable for consideration when addressing service life design. The circles in the fault tree signify issues capable of reducing service life and issues for which the designer needs to develop mitigating strategies. The strategies to address the individual items listed in each circle are provided in Chapter 4 on bridge decks. For most factors listed in the circles, the Guide identifies more than one possible strategy. To complete the process of navigating through the fault tree, all branches appli- cable to the problem need to be considered, and applicable circles checked. Figure 1.20 shows the remainder of the fault tree with the “Due to Loads” branch collapsed for clarity. The decision for selecting the applicable circles is based on specific project con- ditions and requirements. Figure 1.20. Segment of the fault tree with applicable circles checked for the bridge-deck example. Figure 1.20. Segment of the fault tree with applicable circles checked for the bridge-deck example.

36 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 1.11.3 Developing Strategies and Alternative Solutions Once the fault tree is completed and all applicable factors are identified, the individual strategies capable of mitigating the factors can be collected. If software is used to work through the fault tree, then this step can be automated based on the selections made. Table 1.2 summarizes the list of individual strategies capable of mitigating the factors developed for the example bridge-deck problem that are identified as checked circles in Figures 1.19 and 1.20. At this point, the designer has developed a complete list of strategies that can be used for mitigating various factors capable of affecting the service life of the bridge deck that are based on project requirements. Table 1.2 provides sections of the Guide to which the user can refer for more information, as well as listing the advantages and disadvantages of the defined strategies. Some of the strategies may contradict each other, and others may result in similar results. Intentionally, the Guide does not pro- vide a single strategy or attempt to identify the best strategy. In many cases strategies to mitigate the individual factors capable of reducing service life are context sensitive, meaning that the best strategy is very much dependent on such factors as local prac- tice, environment, or owner preferences. As mentioned, some of the strategies may contradict each other and some may be more preferable because of local practices or owner preferences. Consequently, the next step for the designer is to select strategies that are desirable for each factor affect- ing the service life. Table 1.3 shows a narrower list of strategies extracted from the complete list given in Table 1.2. The first row in Table 1.3 shows the applicable factors that can reduce the service life of the bridge deck under consideration. These factors were obtained by navigating through branches of the fault tree. In developing the information shown in Table 1.3, the designer may consider many factors and ensure that there are no contradicting strategies. For instance, appropriate concrete mix is specified as one strategy to address wear, differential shrinkage, freeze– thaw cycles, humidity, ASR, and alkali-carbonate reactivity (ACR). The designer must ensure that a concrete mix capable of addressing all of these issues can be developed. Otherwise, for a particular factor the designer may be forced to use another strategy. For differential shrinkage, for example, a low modulus of elasticity concrete is needed, but for wear, a high modulus of elasticity is needed. Consequently, to address wear and differential shrinkage, the same concrete mix cannot be used to mitigate both factors, and within a given deck alternative, one of these factors should be mitigated using a different strategy. The next step in the process is to develop possible deck alternatives that meet both LRFD specifications and Guide requirements. Using the information provided in Table 1.3 and ensuring there is no contradic- tion among strategies to mitigate various factors, Table 1.4 shows four possible deck alternatives capable of mitigating all factors affecting the service life of the bridge for the example under consideration. The four alternatives shown in Table 1.4 are project- specific solutions. It is also possible to automate this step by first identifying all possible deck alternatives based on all possible combinations of strategies listed in Table 1.3, and eliminating those judged not feasible because of contradiction among strategies.

37 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK tA B LE 1 .2 . Li St o F in di vi du AL S tr At eg ie S cA PA bL e oF m it ig At in g FA ct or S AF Fe ct in g br id ge d ec k Se rv ic e Li Fe F or e xA m PL e Pr ob Le m Se rv ic e Li fe Is su e C o rr es p o n d in g P ro je ct R eq u ir em en ts Se ct io n M it ig at in g S tr at eg y A d va n ta g e D is ad va n ta g e O ve rl o ad H L 93 w ith 2 0- ki p w he el lo ad , a pp lie d on ce a m on th 4. 3. 1. 1 In cr ea se d ec k th ic kn es s. M in im iz es c ra ck in g. A dd s w ei gh t to b rid ge st ru ct ur e, in cr ea se s co st . M in im iz e ba r sp ac in g fo r gi ve n am ou nt o f s te el . Im pr ov es c ra ck c on tr ol . M or e la bo r to in st al l a nd h ig he r co st . Fa ti g u e 24 ,0 00 a ve ra ge d ai ly tr af fic n or th bo un d an d so ut hb ou nd an d 10 % t ru ck v ol um e 4. 3. 1. 1 D es ig n pe r LR FD s pe ci fic at io ns . M in im iz es p os si bi lit y of re in fo rc em en t fa ilu re . M ay in cr ea se a re a of s te el . W ea r an d ab ra si o n St ud de d tir es o n hi gh le ve l o f s er vi ce b rid ge 4. 3. 1. 1 Im pl em en t co nc re te m ix d es ig n st ra te gi es . Id en tifi ed in C ha pt er 3 . Id en tifi ed in C ha pt er 3 . Im pl em en t m em br an es a nd ov er la ys . Pr ot ec ts s ur fa ce fr om d ire ct co nt ac t w ith t ire s. Re qu ire s pe rio di c re ha bi lit at io n ev er y 10 t o 20 y ea rs . D if fe re n ti al sh ri n k ag e N o re qu ire m en ts fo r ex am pl e pr ob le m 4. 3. 1. 1 U se lo w -m od ul us c on cr et e m ix de si gn fo r co m po si te d ec ks . A llo w s ad di tio na l s tr ai n to b e ac co m m od at ed u p to c ra ck in g st re ss . Ty pi ca lly lo w er in s tr en gt h an d m ay b e su bj ec t to w ea r an d ab ra si on . U se h ig h- cr ee p co nc re te m ix de si gn ed fo r co m po si te d ec ks . Re du ce s lo ck ed -in s tr es se s. U nc om m on m ix d es ig n. D iffi cu lt to a ss es s st re ss r el ie f. D ev el op c om po si te a ct io n af te r co nc re te h as h ar de ne d. A llo w s sl ip pa ge b et w ee n de ck an d su pp or tin g m em be rs , m in im iz in g lo ck ed -in s tr es se s. Li tt le e xp er ie nc e w ith ex pe rim en ta l s ys te m s. Fr ic tio n re du ct io n di ffi cu lt to as se ss . I nt ro du ce s nu m er ou s co ns tr uc tio n jo in ts . G ro ut in te gr ity is su es in c lo se d vo id sy st em s. U se p re ca st d ec k pa ne ls . A llo w s sl ip pa ge b et w ee n de ck an d su pp or tin g m em be rs , m in im iz in g lo ck ed -in s tr es se s. In tr od uc es n um er ou s co ns tr uc tio n jo in ts . (c on tin ue d)

38 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Se rv ic e Li fe Is su e C o rr es p o n d in g P ro je ct R eq u ir em en ts Se ct io n M it ig at in g S tr at eg y A d va n ta g e D is ad va n ta g e Sy st em - fr am in g re st ra in t D ec k sh rin ka ge r es tr ai nt fr om s he ar s tu ds 4. 3. 1. 1 D ev el op a cc ur at e sy st em m od el . Id en tifi es d es ig n cr ite ria fo r es ta bl is hi ng s tr es se s. Re st ra in in g fo rc e m ay c au se cr ac ki ng in d ec k. R ef er t o C ha pt er 8 . R ea ct iv e in g re d ie n ts : A SR /A C R Lo ca l a gg re ga te s ar e re ac tiv e 4. 3. 1. 2 U se m at er ia ls a nd m ix d es ig ns th at a re n ot s en si tiv e to ag gr eg at e. Re fe r to C ha pt er 3 . Re fe r to C ha pt er 3 . C o as ta l cl im at e: H u m id it y Re la tiv e hu m id ity av er ag e 70 % 4. 3. 1. 2 U se m at er ia ls t ha t ar e no t se ns iti ve t o m oi st ur e co nt en t. Re fe r to C ha pt er 3 . Re fe r to C ha pt er 3 . T h er m al cl im at e: Fr ee ze –t h aw M ul tip le fr ee ze –t ha w cy cl es e xp ec te d 4. 3. 1. 2 Re fe r to C ha pt er 3 fo r st ra te gi es re la tin g to fr ee ze –t ha w . Re fe r to C ha pt er 3 fo r st ra te gi es re la tin g to fr ee ze –t ha w . Re fe r to C ha pt er 3 fo r st ra te gi es re la tin g to fr ee ze –t ha w . T h er m al cl im at e: D ei ci n g sa lt s Po te nt ia l f or h ig h ch lo rid e co nc en tr at io ns 4. 3. 1. 2 U se im pe rm ea bl e co nc re te . In cr ea se s pa ss iv ity a ro un d re in fo rc em en t. R ef er t o C ha pt er 5 . H ig h in iti al s hr in ka ge , w hi ch ca n re su lt in c ra ck in g. U se c or ro si on -r es is ta nt re in fo rc em en t. El im in at es d ec k sp al ls , de la m in at io ns , a nd c ra ck in g fr om r ei nf or ce m en t. H ig h co st . L im ite d av ai la bi lit y. So m e pe rf or m an ce is su es a s no te d in C ha pt er 3 . U se w at er pr oo f m em br an es o r ov er la ys . M in im iz es in tr us io n of d is so lv ed ch lo rid es in to d ec k. E as ily re ha bi lit at ed . Re qu ire s pe rio di c re ha bi lit at io n ev er y 10 t o 20 y ea rs . U se e xt er na l p ro te ct io n m et ho ds , s uc h as c at ho di c pr ot ec tio n. Re du ce s co rr os io n. R ef er t o C ha pt er 5 . H ig h co st . R eq ui re s ex te ns iv e m ai nt en an ce a nd a no de /b at te ry . U se e ffe ct iv e dr ai na ge t o ke ep su rf ac e dr y, m in im iz e po nd in g. M in im iz es in tr us io n of d is so lv ed ch lo rid es in to d ec k. Re qu ire s m ai nt en an ce o f dr ai na ge . U se p er io di c pr es su re w as hi ng to r em ov e co nt am in an ts . M in im iz es in tr us io n of d is so lv ed ch lo rid es in to d ec k. L ow c os t. Re qu ire s de di ca te d m ai nt en an ce st af f a nd a pp ro pr ia te b ud ge t. U se n on ch lo rid e- ba se d de ic in g so lu tio n. El im in at es c or ro si on fr om ch lo rid es . H ig h co st . tA B LE 1 .2 . Li St o F in di vi du AL S tr At eg ie S cA PA bL e oF m it ig At in g FA ct or S AF Fe ct in g br id ge d ec k Se rv ic e Li Fe Fo r ex Am PL e Pr ob Le m (­c on tin ue d) (c on tin ue d)

39 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK Se rv ic e Li fe Is su e C o rr es p o n d in g P ro je ct R eq u ir em en ts Se ct io n M it ig at in g S tr at eg y A d va n ta g e D is ad va n ta g e C o as ta l cl im at e: Sa lt s p ra y Sp la sh p ot en tia l f ro m je t sk is ( ro os te rt ai ls ) 4. 3. 1. 2 U se im pe rm ea bl e co nc re te . In cr ea se s pa ss iv ity a ro un d re in fo rc em en t. R ef er t o C ha pt er 5 . Ty pi ca lly lo w er in s tr en gt h an d m ay b e su bj ec t to w ea r an d ab ra si on . U se c or ro si on -r es is ta nt re in fo rc em en t. El im in at es d ec k sp al ls , de la m in at io ns , a nd c ra ck in g fr om r ei nf or ce m en t co rr os io n. Re fe r to C ha pt er 3 . H ig h co st . L im ite d av ai la bi lit y. So m e pe rf or m an ce is su es a s no te d in C ha pt er 3 . U se w at er pr oo f m em br an es / ov er la ys o n tr av el s er vi ce s of br id ge d ec k. M in im iz es in tr us io n of d is so lv ed ch lo rid es in to d ec k. Re qu ire s pe rio di c re ha bi lit at io n ev er y 10 t o 20 y ea rs . U se e xt er na l p ro te ct io n m et ho ds , s uc h as c at ho di c pr ot ec tio n. Re du ce s co rr os io n. R ef er t o C ha pt er 5 . H ig h co st . R eq ui re s ex te ns iv e m ai nt en an ce a nd a no de / ba tt er y. U se s ea le rs o n no nt ra ve l s ur fa ce s of b rid ge d ec k. M in im iz es in tr us io n of d is so lv ed ch lo rid es in to d ec k. Re qu ire s pe rio di c re ha bi lit at io n ev er y 5 to 1 0 ye ar s. U se c or ro si on -r es is ta nt s ta y- in -p la ce fo rm s on b ot to m o f br id ge d ec k. M in im iz es in tr us io n of d is so lv ed ch lo rid es in to d ec k. D iffi cu lt to in sp ec t. U se e ffe ct iv e dr ai na ge t o ke ep su rf ac e dr y. M in im iz es in tr us io n of d is so lv ed ch lo rid es in to d ec k. Re qu ire s m ai nt en an ce o f dr ai na ge a nd p er io di c cl ea ni ng . U se p er io di c pr es su re w as hi ng to r em ov e co nt am in an ts . M in im iz es in tr us io n of d is so lv ed ch lo rid es in to d ec k. Re qu ire s de di ca te d m ai nt en an ce s ta ff an d ap pr op ria te b ud ge t. N ot e: A C R = al ka li- ca rb on at e re ac tiv ity ; A SR = a lk al i-s ili ca r ea ct iv ity . tA B LE 1 .2 . Li St o F in di vi du AL S tr At eg ie S cA PA bL e oF m it ig At in g FA ct or S AF Fe ct in g br id ge d ec k Se rv ic e Li Fe Fo r ex Am PL e Pr ob Le m (­c on tin ue d)

40 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE tABLE 1.3. LiSt oF StrAtegieS SPeciFic For deveLoPing deck ALternAtiveS Issue Strategy Overload Increase deck thinness Fatigue Design per AASHTO Wear Concrete mix Membrane and overlay Increase thickness System restraint Accurate modeling during structural analysis of bridge system Differential shrinkage Concrete mix: Use mix with low modulus Deicing Impermeable concrete Stainless steel Specify nonchloride-based deicing Membrane and overlay Freeze–thaw Concrete mix: Air content Salt spray Stainless steel Stay-in-place metal deck to protect bottom Deck bottom sealer and top membrane Humidity Use aggregate that is not sensitive to humidity ASR/ACR Concrete mix nonreactive aggregate For each alternative shown in Table 1.4, Rows 2 through 11 show the service life design factors identified in Table 1.3 and corresponding strategies selected for each alternative. Incorporating all of the select strategies listed in Rows 2 through 11 for each of the four alternatives results in the modified deck configurations shown in the bottom row of Table 1.4. Development of the four deck alternatives signifies the com- pletion of Step 4a from Figure 1.18. Although the strategies listed in Table 1.3 could lead to the development of additional deck alternatives, for the sake of simplicity only four alternatives are shown in Table 1.4. Alternative 1 in Table 1.4 represents a design that meets the strength requirements stated in LRFD specifications. The total deck thickness is 8 in., with no consideration for any of the factors capable of reducing service life. The main feature of Alternative 1 is having impermeable concrete, with 5% silica fume and 10% fly ash. The addition of fly ash is assumed to affect the rate of reduction in the diffusivity of concrete, a parameter used in estimating the time to initiate corrosion. Alternative 2 in Table 1.4 relies mainly on the use of stainless steel reinforcement, in this case Grade 316 stainless steel, to prevent corrosion. Alternative 3 in Table 1.4 uses regular concrete with increased cover to delay the time to initiate the corrosion, and Alternative 4 uses a membrane and overlay to address corrosion. All alternatives use increased thickness to address overload, increasing the deck thickness by 0.5 in. Table 1.4 also presents additional strategies to address factors capable of reducing service life.

41 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK tA B LE 1 .4 . de ve Lo Pi n g de ck A Lt er n At iv eS m ee ti n g bo th A AS ht o An d Gu id e re Qu ir em en tS Is su e C o n fi g u ra ti o n p er A A SH T O D es ig n R eq u ir em en ts A lt er n at iv e 1 A lt er n at iv e 2 A lt er n at iv e 3 A lt er n at iv e 4 O ve rlo ad Ba se d es ig n In cr ea se t hi ck ne ss b y 0. 5 in . In cr ea se t hi ck ne ss b y 0. 5 in . In cr ea se t hi ck ne ss b y 0. 5 in . In cr ea se t hi ck ne ss b y 0. 5 in . Fa tig ue Ba se d es ig n D es ig n pe r A A SH TO . D es ig n pe r A A SH TO . D es ig n pe r A A SH TO . D es ig n pe r A A SH TO . W ea r Ba se d es ig n In cr ea se t hi ck ne ss b y 0. 5 in . In cr ea se t hi ck ne ss b y 0. 5 in . In cr ea se t hi ck ne ss b y 0. 5 in . M em br an e an d ov er la y. Sy st em r es tr ai nt Ba se d es ig n A cc ur at e m od el in g du rin g an al ys is o f t he sy st em . A cc ur at e m od el in g du rin g an al ys is o f t he sy st em . A cc ur at e m od el in g du rin g an al ys is o f t he sy st em . A cc ur at e m od el in g du rin g an al ys is o f t he sy st em . D iff er en tia l sh rin ka ge Ba se d es ig n C on cr et e m ix : U se m ix w ith lo w m od ul us . C on cr et e m ix : U se m ix w ith lo w m od ul us . C on cr et e m ix : U se m ix w ith lo w m od ul us . C on cr et e m ix : U se m ix w ith lo w m od ul us . D ei ci ng Ba se d es ig n Im pe rm ea bl e co nc re te : 5% s ili ca fu m e, 1 0% fl y as h. St ai nl es s st ee l re in fo rc em en t. In cr ea se c ov er b y 1 in . M em br an e an d ov er la y. Fr ee ze –t ha w Ba se d es ig n C on cr et e m ix : A ir co nt en t. C on cr et e m ix : A ir co nt en t. C on cr et e m ix : A ir co nt en t. C on cr et e m ix : A ir co nt en t an d m em br an e an d ov er la y. Sa lt sp ra y Ba se d es ig n Im pe rm ea bl e co nc re te : 5% s ili ca fu m e, 1 0% fl y as h. St ai nl es s st ee l re in fo rc em en t. Se al t he b ot to m u si ng st ay -in -p la ce m et al d ec k; to p is p ro te ct ed b y in cr ea si ng c ov er . Se al t he b ot to m u si ng st ay -in -p la ce m et al d ec k; to p is p ro te ct ed b y m em br an e an d ov er la y. C oa st al : H um id ity Ba se d es ig n U se a gg re ga te s th at a re no t se ns iti ve t o hu m id ity . U se a gg re ga te t ha t ar e no t se ns iti ve t o hu m id ity . U se a gg re ga te t ha t ar e no t se ns iti ve t o hu m id ity . U se a gg re ga te t ha t ar e no t se ns iti ve t o hu m id ity . A SR a nd A C R Ba se d es ig n C on cr et e m ix : N on re ac tiv e ag gr eg at e. C on cr et e m ix : N on re ac tiv e ag gr eg at e. C on cr et e m ix : N on re ac tiv e ag gr eg at e. C on cr et e m ix : N on re ac tiv e ag gr eg at e. St ra te gy A s de si gn ed (L RF D s tr en gt h) A s de si gn ed w ith th ic ke ne d de ck a nd im pe rm ea bl e co nc re te . A s de si gn ed w ith th ic ke ne d de ck an d st ai nl es s st ee l re in fo rc em en t. A s de si gn ed w ith th ic ke ne d de ck , b la ck st ee l r ei nf or ce m en t, d ec k bo tt om s ea le d w ith s ta y- in -p la ce fo rm . A s de si gn ed w ith b la ck st ee l r ei nf or ce m en t, d ec k bo tt om s ea le d by m et al de ck a nd t op o f d ec k m em br an e. C on fig ur at io n

42 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 1.11.4 Evaluating Alternatives The next step is predicting the service life of each alternative (Step 5 in Figure 1.17) and comparing it with the design service life of the bridge system as specified by the owner and project requirements. Based on the outcome, the development of rehabili- tation or replacement requirements (Step 5a in Figure 1.17) and a maintenance plan (Step 6 in Figure 1.17) may be necessary. The last step for the bridge-deck alternative under consideration (Step 7 in Figure 1.17) is to perform LCCA for comparison. As described in Chapter 2 on system selection, the designer must also consider the interactions that might exist among various parts of the bridge. This step is not covered for this example. The potential service life of each deck alternative can be calculated based on the assumption that the main mode of deterioration is ingress of chloride into concrete, which can result in corrosion of reinforcement. One approach is to use the solution to Fick’s second law as shown in the following equation. In a one-dimensional case, Fick’s law can be expressed and illustrated as shown by Equation 1.5: C C x D t 1 erf 2x t c , 0= −      ( ) (1.5) where C(x,t) = chloride concentration at depth x and time t, C0 = surface chloride concentration (kg/m 3 or lb/yd3), Dc = chloride diffusion constant (cm 2/year or in.2/year), and erf = error function (from standard mathematical tables). The use of Fick’s law to determine the time of corrosion initiation is described in Chapter 5, Section 2 of the Guide. Equation 1.5 can be used to assess ingress of chloride through the concrete cover. As an example, Figure 1.21 indicates the type of information that can be developed, which shows chloride concentration through the deck thickness for three time periods after a deck is cast. The information shown in Figure 1.21 can be used to predict the time when corro- sion will be initiated, which in turn can be used to estimate the service life of the bridge deck if corrosion of reinforcement is the main mode of deterioration. The Fick’s law in this case is the deterioration model. To complete the example, Life-365, a free program developed by the concrete industry, is used to conduct the LCCA and assist in selecting an optimum solution. Life-365 uses a finite-difference approach to solve Fick’s second law and to estimate the time to initiation of corrosion. Other approaches, such as an error function solu- tion to Fick’s second law, Equation 1.5, could also be used. The solution to Fick’s second law estimates the time to initiation of corrosion (t i). For the example, it is assumed (assumption within Life-365) that once the corrosion is initiated, the time to propagate the corrosion to the point at which repair is needed (tr) is a constant 6 years, regardless of concrete mix used. After the time period ti + tr,

43 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK Life-365 assumes that repair action is at set time intervals, say, every 10 or 20 years, and set cost per unit area in square feet. Further, within each repair cycle, it is assumed that only a portion of the deck area will be repaired. For instance, within each repair cycle, only 10% or 20% of the deck will need repair. The time to initiate corrosion depends on concrete mix and preventive measures, such as use of stainless steel, concrete cover, or membranes. Life-365 follows the guid- ance and terminology in ASTM E-917, Standard Practice for Measuring the Life-Cycle Costs of Buildings and Building Systems. The final number that can be used to select the optimal deck alternative can be the life-cycle cost, which is the initial cost plus the present value of all future rehabilitation costs over the desired service life, in this case 100 years. Table 1.5 shows the input parameters used within Life-365 to conduct an LCCA for each of the four alternatives shown in Table 1.4. It is assumed that the bridge is located in Boston, Massachusetts; has a required service life of the deck of 100 years; and, for the sake of comparison, a total surface area of the deck of 10,000 square ft. Table 1.5 also shows the yearly temperature pro- file used. The diffusion coefficient and ingress of chloride are influenced by tempera- ture fluctuation. The input parameters shown in Table 1.5 are applicable to all four alternatives shown in Table 1.4. The specific input and end results for each alternative are shown in Table 1.6. As indicated in Table 1.6, Alternatives 1, 2, 3, and 4 use the same concrete mix, referred to here as regular mix, which was used for the base option designed in accor- dance with LRFD specifications (this option is designated as AASHTO Design in Table 1.4). Alternative 1 uses 5% silica fume to make the concrete impermeable. Alter- native 2 uses stainless steel. Alternative 3 uses increased concrete cover to delay the adverse effects of corrosion, and Alternative 4 uses a membrane and overlay. A brief description of the results for each alternative follows. Deck Design Based on LRFD Specifications. As shown in Table 1.6, in the case of the AASHTO base design, corrosion starts after 6.6 years (shown as Initiation). Thereafter, the propagation of corrosion to the point of needed repair is 6 years. At the end of 12.6 years, repair and maintenance actions are assumed to start and continue Figure 1.21. Chloride concentration within concrete over time.

44 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE every 10 years, during which 20% of the surface area is repaired. These assumptions are used for the sake of demonstration and will vary based on various DOT prefer- ences and practices. Figure 1.22 shows the total life-cycle cost based on present value. The total life-cycle cost for the AASHTO base design is $774,676, which is shown in the bottom row of Table 1.6. An inflation rate of 1.6% and a discount rate of 2% were used in developing the total life-cycle cost. As indicated in Table 1.6, the initial cost of the AASHTO base design is the lowest of the alternatives ($37, 215). However, the total life-cycle cost is the highest ($774,676). It should be mentioned that these LCCAs ignore the user costs, or any other cost to society during repair and closure to traffic, and these costs can be significant. tABLE 1.5. inPut PArAmeterS in LiFe-365 For Four ALternAtiveS And yeArLy temPerAture ProFiLe Parameter Value Base units U.S. units Concentration units Weight concentration (%) Type of structure Slabs and walls (1-D) Third dimension (ft2) 10,000 Base year 2011 Study period (years) 100 Inflation rate (%) 1.6 Discount rate (%) 2 Location Massachusetts Sublocation Boston Exposure type Urban highway bridges Maximum surface concentration (% weight of concrete) 0.68 Time to buildup (years) 7.1 Month Temperature Profile (°F) January –1.9 February –0.9 March 3.7 April 8.9 May 14.6 June 19.8 July 23.1 August 22.2 September 18.2 October 12.7 November 7.4 December 0.9

45 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK tABLE 1.6 PArAmeterS SPeciFic to eAch ALternAtive From tAbLe 1.4 Analysis Parameter AASHTO Design Alternative 1 Alternative 2 Alternative 3 Alternative 4 Concrete mix type Regular Silica fume Regular Regular Regular Water-cement ratio 0.42 0.35 0.42 0.42 0.42 Slag (%) 0 0 0 0 0 Fly ash (%) 0 10 0 0 0 Silica fume (%) 0 5 0 0 0 Steel type Black steel Black steel 316 Stainless Black steel Black steel Steel (%) 1.2 1.2 1.2 1.2 1.2 Propagation period (years) 6 6 6 6 6 Inhibitor none none none none none Barrier none none none none Membrane D28 (in. × in./s) 1.38E-08 4.09E-09 1.38E-08 1.38E-08 1.38E-08 m (diffusion decay coefficient) 0.2 0.28 0.2 0.2 0.2 Initiation (years) 6.6 34.8 100 15.2 22.4 Propagation (years) 6 6 6 6 6 Service life (years) 12.6 40.8 106 21.2 28.4 Use user mix cost? (true or false) False False False False False User mix cost ($/yd3) 0 0 0 0 0 Depth (in.) 8 9 9 10 8.5 Depth to reinforcement (in.) 2 2.5 2.5 3.5 2 Unit Cost Area to repair (%) 20 10 10 20 5 Repair cost ($/ft2) 50 50 50 50 20 Repair interval (years) 10 10 10 10 10 Base mix cost ($/yd3) 80 90 80 80 80 Black steel cost ($/lb) 0.45 0.45 0.45 0.45 0.45 Epoxy steel cost ($/lb) 0.6 0.6 0.6 0.6 0.6 Stainless steel cost ($/lb) 2.99 2.99 2.99 2.99 2.99 Inhibitor cost ($/lb) 5.68 5.68 5.68 5.68 5.68 Membrane cost ($/ft2) 7 7 7 7 7 Sealant cost ($/ft2) 0.65 0.65 0.65 0.65 0.65 Result Repair interval (years) 10 10 none 10 10 Base cost ($) 37,215 44,645 152,753 46,519 39,541 Barrier cost ($) 0 0 0 0 70,000 Repair cost ($) 737,461 232,905 0 644,595 62,711 Life-cycle cost ($) 774,676 277,550 152,753 691,114 172,252

46 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Alternative 1. This alternative uses impermeable concrete by incorporating 5% silica fume into the mix. As a result, the initial unit cost of concrete is assumed to increase by $10/yd3. The initiation of corrosion starts at 34.8 years after casting, and the propagation phase lasts 6 years. Therefore, after 40.8 years the repair procedure is assumed to start. It is assumed that the repair procedure is to be conducted every 10 years and further assumed that during each repair cycle 10% of the surface area is to need repair. Other repair alternatives can consist of complete replacement of the deck every 40 years. The total life-cycle cost of Alternative 1 for the assumptions made is $277,550 (Table 1.6). The initial cost of Alternative 1 ($44,645) is slightly higher than the initial cost for the AASHTO base design ($37,215). Alternative 2. This alternative uses Grade 316 stainless steel to address the issue of reinforcement corrosion. The time to initiation of corrosion for this alternative is more than 100 years, and consequently no repair action is needed. The initial cost of using stainless steel is the highest among all alternatives ($152,753). However, the total life- cycle cost associated with Alternative 2 ($152,753 as shown in Table 1.6) is the lowest among all the alternatives. Alternative 3. This alternative uses increased cover to delay the initiation of cor- rosion. Using increased cover delays the corrosion initiation from 6.6 years for the AASHTO base design to 15.2 years (Table 1.6). The total cost of Alternative 3 is rela- tively high, $691,114. There does not seem to be much benefit in using this alternative, especially considering that increasing the concrete cover will subject the substructure and foundations to higher dead loads. 0 10 20 30 40 50 60 70 80 90 2005 2015 2025 2035 2045 2055 2065 2075 2085 2095 2105 2115 C on st an t D ol la rs ($ p er sq . f t) Year Cumulative Present Value Base Case Figure 1.22. Total life-cycle cost for LRFD specifications–based design.

47 Chapter 1. DESiGN FOR SERviCE LiFE: GENERAL FRAMEWORK Alternative 4. Alternative 4 uses a membrane and overlay to prevent corrosion of reinforcement. The total deck thickness is only 8.5 in. compared with 9 in. and 10 in. for Alternatives 1, 2, and 3. The repair cost per square foot for this alternative is assumed to be lower at $20/ft2 as compared with $50/ft2 for others when a membrane is used. It is assumed that a high-quality membrane at $7/ft2 is used and that it will last 75 years. Consequently, the repair will involve replacing damaged overlay areas, which are assumed to be 5% of the total surface area during each repair cycle—every 10 years starting 28.4 years after the initial installation (Table 1.6). At $172,252, the total life-cycle cost of this alternative, using membrane, is very low, but the initial cost ($39,541 + $70,000) is more than twice the AASHTO base design cost of $37,215. The use of a membrane could be much more economical than that indicated by this example. For instance, the calculation leads to the conclusion that corrosion will start after 28.4 years, which is not realistic. The concrete deck below the membrane could last a long time without any need for repair, and any needed repair action would only be for replacing the thin overlay, which could be achieved quickly with minimal inter- ruption to traffic. These factors were not considered in conducting the LCCA for this alternative. 1.11.5 Summary and Conclusion Table 1.7 summarizes the results for all alternatives. Using this information, it is fea- sible to conclude that the use of stainless steel or membrane plus overlay can provide the best economy. 1.12 Future deveLoPment oF the Guide The Guide provides a general, comprehensive framework for designing new bridges and rehabilitating existing bridges for service life. The approach presented by the Guide is flexible and can be adapted as new information becomes available. The Guide also provides a platform for developing customized manuals by state DOTs or for developing a customized and systematic approach for the service life design of major and complex bridges. tABLE 1.7. ALternAtive SummAry Alternative Main Feature to Address Corrosion Initial Cost Life-Cycle Cost AASHTO base design Not applicable $37,215 $774,676 1 Impermeable concrete using silica fume $44,645 $277,550 2 Use of 316-stainless steel $152,753 $152,753 3 Increasing concrete cover $46,519 $691,114 4 Using membrane and overlay $109,541 $172,252

48 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Design for service life is a context-sensitive problem: local agency practices and preferences are important. Customizing the Guide can be achieved by using the gen- eral framework outlined in these pages and incorporating strategies and solutions pre- ferred by each DOT for factors affecting the service life of its bridges. One of the challenges in developing true service life design is the lack of reliable, available deterioration models that are based on either field data or laws of physics governing the deterioration. Several studies are under way to develop deterioration models for various bridge elements, components, and subsystems. Such information can be incorporated into the Guide as it becomes available. These models are needed to further develop reliable LCCAs. There is a need to develop specific LCCA tools dedicated to bridges, with the ability to incorporate user costs when applicable. These tools must be flexible enough to allow incorporating new information and deterioration models as they become available. There is a further need to develop more comprehensive examples that take into account the interaction between solutions that may seem appropriate for an individual bridge element, component, or subsystem when viewed in isolation, and yet are less than optimum when considering service life solutions for the combined bridge system. Finally, the significant amount of information provided in the Guide is time consuming to comprehend in its entirety. There is a need to automate the use of the Guide by developing tools that would facilitate navigating through all the included information.