Performance of Bridges That Received Funding Under the Innovative Bridge Research and Construction Program (2019)

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.

35 4 Performance of the IBRC Bridges; Utility of the IBRC Technologies The following two sections present evidence relevant to the second task in the committee’s charge: to analyze the utility, compared to conventional materials and technologies, of each of the innovative materials and tech- nologies used in projects for bridges under the program in meeting needs for a sustainable and low life-cycle cost transportation system. The evidence comes from two types of sources: evaluations in the published engineer- ing literature of the technologies used in Innovative Bridge Research and Construction (IBRC) projects and data from the 10 case study state trans- portation departments on the performance of their IBRC bridges. The com- mittee’s conclusions on the utility of the IBRC technologies are presented in Chapter 5. QUANTITATIVE EVALUATIONS OF THE BENEFITS OF IBRC TECHNOLOGIES Most of the IBRC projects have not been systematically monitored for the purpose of evaluating the performance of the innovative technologies that they demonstrated. Therefore, the evidence available from the IBRC proj- ects themselves on the value of the technologies is limited. However, many of the IBRC technologies are widely used, and evaluations have been pub- lished of their performance and costs in projects other than IBRC projects. Summarized in the following section are selected published evalua- tions of in-service performance and life-cycle cost comparisons for the major categories of IBRC technologies: advanced concrete materials, fiber- reinforced polymer (FRP) composite materials, corrosion control

36 PERFORMANCE OF BRIDGES technologies, high-performance steel (HPS), and accelerated bridge con- struction (ABC). The committee did not conduct a comprehensive litera- ture review of performance evaluations of the technologies. The studies cited are representative of the literature and provide a part of the basis for the committee’s conclusions on the utility of the technologies. The content of the studies reviewed suggests that overall, evaluation of the long-term performance of the IBRC technologies has been fragmen- tary. Life-cycle cost comparisons of innovative technologies are based on projections of future performance (that is, assuming that the innovative technology will perform as intended) rather than on actual past experience. Systematic long-term monitoring of the durability of IBRC materials and technologies in bridge projects has rarely been conducted. Advanced Concrete Materials The IBRC technologies in this category are high-performance concrete (HPC), self-consolidating concrete (SCC), and ultra-high performance con- crete (UHPC). HPC HPC was developed under the first Strategic Highway Research Program (SHRP) implementation in the early to mid-1990s for use in all bridge ele- ments (Halladay 1998). Concrete mixtures, concrete properties, research projects, girder fabrication, bridge construction, live-load tests, and speci- fications from 19 HPC bridges in 14 states were compiled in 2006 to docu- ment SHRP implementation (Russell et al. 2006). High-strength HPC has been used successfully in bridge girders for many years. Although high-strength HPC has been used in some bridge decks and substructures, the preference for decks and substructures is typically to specify normal-strength HPC. This is because high compressive strength is typically not required in decks and substructures and the high cementitious content required for high strength can lead to an increased potential for cracking. A synthesis of concrete bridge deck performance was conducted in 2004. Findings included that all HPC is not high-strength concrete and experience has shown that the use of high-strength concrete does not nec- essarily lead to a highly durable concrete or, conversely, a highly durable concrete is not necessarily a high-strength concrete. Research and practice show that designing for durability involves more than specifying compres- sive strength. The researcher identified parameters, based on current prac- tice and research results, that enhance the performance of concrete decks.

PERFORMANCE OF THE IBRC BRIDGES 37 These include specified fly ash, silica fume, and ground-granulated blast furnace slag replacement as percentages of the total cementitious materials content; maximum water–cementitious materials ratio; maximum concrete permeability; and 6-ksi maximum concrete compressive strength (Russell 2004, 13, 29). In 2006 a life-cycle cost analysis was performed on two concrete high- way bridge decks built in a corrosive environment using HPC and conven- tional concrete. The analysis showed an estimated service life of the HPC deck from 3 to 10 times the life of the conventional deck. In addition, the HPC deck was found to be more cost-effective than the conventional deck, with agency life-cycle cost about 40 to 45 percent lower and the user’s life- cycle cost about one-third the cost of the conventional deck (Daigle and Lounis 2006). SCC Extensive research in the past 20 years indicates important benefits from the use of SCC in bridge construction (Bailey at al. 2005; HDR 2012; Henault 2014; Ozyildirim 2008). In areas of the country where concrete suppliers and contractors were unfamiliar with SCC, there were some less than desir- able outcomes. Yet, experienced suppliers and contractors are consistently delivering structures with enhanced service life and life-cycle costs through the use of SCC. UHPC The enhanced durability of UHPC compared to conventional concrete is expected to result in structures with a longer service life and reduced main- tenance needs compared to those constructed with conventional concrete, and thus reduced life-cycle costs. Piotrowski and Schmidt (2012) conducted a life-cycle cost analysis of two replacement methods for the Eder Bridge in Felsberg, Germany. One used precast UHPC box girders filled with lightweight concrete and the other used conventional prestressed concrete bridge girders. The UHPC bridge with the higher initial costs was predicted to have a lower life-cycle cost over 100 years. The 2013 Federal Highway Administra tion (FHWA) UHPC state-of-the-art report found that research had not yet been conducted to demonstrate that cost savings from the greater durability of UHPC compared with conventional concrete will be sufficient to offset higher initial cost of the material and thus reduce life- cycle cost. The report recommended research on the cost-effectiveness of UHPC in various applications (Russell and Graybeal 2013, 67–69).

38 PERFORMANCE OF BRIDGES FRP Composite Technology The technologies in this category are externally bonded FRP reinforcement; FRP deck elements; FRP beams, girders, and appurtenances; FRP rebar; and FRP prestressing tendons (strand or bar). A National Institute of Standards and Technology study used life- cycle cost comparisons of three FRP composite bridge deck designs and a conventional concrete deck to illustrate a proposed standard method for life-cycle cost evaluation of new materials and designs (Ehlen and Marshall 1996). The results indicated that “new technology introduction” costs, the extra time and labor required to design and monitor a project involving a new technology, could negate the long-term cost savings obtainable from FRP decks in initial projects, but once the new technology introduction costs are spread over several projects, they become negligible (Ehlen and Marshall 1996, 44–45). More recent investigations of FRP bridge deck panels (Hastak et al. 2004) and FRP bridge superstructure elements (Eamon et al. 2012) have concluded that life-cycle cost savings are to be expected, compared with conventional construction, because savings in operating costs and from longer lifetimes outweigh higher construction costs. Theoretical and limited case study investigations have concluded that construction with FRP has higher initial cost than construction with con- ventional materials (Alampalli et al. 2002; Soroushian et al. 2001), partially because of material costs and partially because of the experimental nature of the product, although the use of the materials for a bridge superstruc- ture offers an advantage of faster construction compared with concrete (Alampalli et al. 2002). Reports from the IBRC projects indicate that these theoretical studies on costs are substantiated, although in the case of FRP externally bonded reinforcement, there could be an immediate cost savings (Harichandran and Baiyasi 2000). Available field data are too limited to support estimates based on field experience of the comparative life-cycle costs of using FRP materials, either as rebar or decks during construction or in externally bonded reinforcement for repair. Corrosion Control Technologies: Concrete Reinforcement The technologies in this category include three types of concrete rebar: solid and clad stainless steel, low-chromium, and galvanized. Coating reinforcement steel with epoxy became mainstream in the 1980s to extend the service lives of highway structures exposed to chlorides from deicing chemicals and spray and splash from saltwater. Research on epoxy coating suggests it can add 5 to 15 years of service life compared to the use of bare steel (Kahl 2007). Corrosion-resistant rebar materials are

PERFORMANCE OF THE IBRC BRIDGES 39 intended to add even more years of service life. The technologies in this category that were examined in the IBRC program include three types: low- chromium, galvanized, and solid and clad stainless steel rebar. Some research results in the literature suggest the service life of rein- forced concrete can be increased by the use of corrosion-resistant reinforc- ing steel. The Virginia Transportation Research Council (VTRC), which is affiliated with the Virginia Department of Transportation (VDOT), issued reports in 2007 and 2018 that examined the effect that cracks in bridge decks can have on chloride penetration and the onset of rebar corrosion (Williamson et al. 2007; Balakumaran et al. 2018). VTRC conducted a lit- erature review and studied 37 highway bridge decks. Ten of the decks were older (built from 1968 to 1971) and built with uncoated rebar. The other 27 were built with epoxy-coated rebar from 1984 to 1991. Because all 37 decks predated VDOT’s use of corrosion-resistant rebar (which began in the 2000s), the ability of rebar technologies to resist corrosion under different chloride exposures had to be modeled based on the study results. Service life estimates from VTRC’s modeling suggest that when decks have low to medium cracking frequencies, both low-chromium and stainless steel rebar offer good corrosion resistance. With medium crack frequencies, low-chromium steel rebar was estimated to resist corrosion for more than 50 years. However, with high crack frequencies that allow higher chloride diffusion under heavy use of deicing chemicals, the time to corrosion onset in low-chromium steel was estimated to be as soon as 30 years. In the case of stainless steel, both the frequency of cracking and degree of chloride dif- fusion through the cracks had little effect on corrosion rates, as service life was estimated to exceed 150 years under all circumstances. The report con- cluded that because repairing cracks can be expensive, corrosion-resistant technologies such as low-chromium and stainless steel rebar can be cost- effective on a life-cycle basis under conditions of high chloride exposure. VTRC prepared a 100-year life-cycle cost comparison of epoxy-coated rebar and the two types of corrosion-resistant rebar, as shown in Table 4-1. Based on its modeling and its life-cycle estimates—which assume a re- duced need for crack sealing and overlays and patching of concrete decks with corrosion-resistant rebar—VTRC recommended that VDOT under- take follow-up validation studies of newer bridge decks built with low- chromium and stainless steel rebar.1 1 Table 4-1 shows undiscounted future expenditures. If future expenditures are discounted at a rate of 3 percent per year, the present values of the 100-year life cost, including patching and user costs are $560 per yd2 for epoxy coated, $426 for low-chromium, and $483 for stainless steel. If expenditures are discounted at 0.6 percent, the 2018 rate specified by the Office of Management and Budget for discounting constant-dollar expenditures over periods of 30 years or longer in evaluations of federal government programs (OMB 2018), the present values are $915 per yd2 for epoxy coated, $631 for low-chromium, and $483 for stainless steel.

40 PERFORMANCE OF BRIDGES TABLE 4-1 Life-Cycle Cost Analysis of Epoxy-Coated and Corrosion- Resistant Rebar Rebar Type Epoxy Coated ASTM A1035 (low-chromium steel) ASTM A955 (stainless steel) Rebar cost, $/lb 1.20 1.65 3.50 Rebar construction cost, $/yd2 72 99 210 Deck construction cost, $/yd2 345 372 483 Seal cracks in deck with ECR, $/yd2 45 — — Polymer overlay @ 20 years, $/yd2 60 — — Concrete overlay @ 40 years, $/yd2 150 — — Polymer overlay @ 50 years, $/yd2 — 60 — Concrete overlay @ 70 years, $/yd2 150 150 — 100-year life cost, $/yd2 750 582 483 Patch overlay @ 60 years, $/yd2 30 — — Patch overlay @ 90 years, $/yd2 30 30 — 100-year life cost, including patching concrete overlays, $/yd2 810 612 483 User cost for concrete overlay @ 40 year, $/yd2 150 — — User cost for concrete overlay @ 70 years, $/yd2 150 150 — 100-year life cost, including patching concrete overlays and user costs for overlays, $/yd2 1,110 762 483 Assumptions: 60 lb of reinforcement per yd2 of deck. Costs based on 2009 and 2010 (through August) bid tabs. A4 concrete @ $852.24 per cubic yard = 213.06 per yd2. Mobilization for deck construction @ $50.00 per yd2. Saw cut grooves @ $10.00 per yd2. Cracks are linear and 9-ft apart = 1 feet of crack per yd2 of deck @ $45.36 per ft. ASTM 1035 reinforcement time to corrosion is four times that of epoxy-coated rebar (ECR). Solid stainless and stainless last more than 100 years. User cost for polymer overlay and patching concrete overlay after 20 years is zero. User cost for concrete overlay equals cost of overlay. SOURCE: Provided to the committee by Michael Sprinkel, Associate Director, VTRC.

PERFORMANCE OF THE IBRC BRIDGES 41 Another life-cycle cost comparison of low-chromium steel rebar was conducted by the Michigan Department of Transportation (Kahl 2007). The investigator concluded that low-chromium steel rebar exhibits corro- sion resistance, higher yield strength, and a lower life-cycle cost than epoxy- coated rebar. The low-chromium steel rebar was estimated to provide an additional 12 years of service life over epoxy-coated rebar, which could justify its higher initial investment under some applications. The investiga- tor concluded that low-chromium steel rebar may be justified on a life-cycle basis when applied on high-volume bridges where service disruptions from deck repairs can be costly. Corrosion Control Technologies: Coating and Anodes The main IBRC corrosion control technologies used in bridge repair are metallizing (coatings) and galvanic protection (sacrificial anodes). The use of metallizing and sacrificial anodes as forms of corrosion control for highway structures was pioneered in Florida in the 1980s to protect bridges in salt- laden marine environments prone to concrete deterioration from corroded reinforcement. Florida’s applications and follow-on research have focused primarily on thermally applied zinc coatings (metallizing) and anode sys- tems (jacketed and point-specific) using zinc and aluminum (Larson 2018; Troconis de Rincon et al. 2018). Florida’s high environmental humidity and saltwater-exposed bridges made the anode technologies more effective. The reduced resistivity due to moisture in the concrete and the high rate of oxygen diffusion in the splash zone resulted in higher passivating effects of the cathodic currents. Florida also evaluated bridges that have metallizing systems and humectants, which are substances that assist in the retention of humidity at the interface of the concrete. Various types of humectants were studied. Lithium-salt-based ones were found to perform particularly well by keeping humidity high at the interface to enable a higher current in the zinc and more effective corrosion protection as a result. Likewise, a study conducted in Australia evaluated sacrificial anodic protection systems as a corrosion control measure for bridge decks in coastal environments (Moore et al. 2012). As in Florida, the study indicated good success with sacrificial anodic systems, especially zinc strip anodes. It was noted that sacrificial anode systems were particularly cost-effective on smaller structures or application areas. Australia’s and Florida’s metallizing and anodic technologies were es- sentially the same ones investigated in the IBRC program; however, their experience is primarily applicable to bridges in marine environments under high humidity conditions. Most of the applications in the IBRC program did not involve bridges in marine environments or bridges exposed to high levels of humidity. The literature contains few studies of these technologies

42 PERFORMANCE OF BRIDGES when used outside marine environments. Because the principal application of coatings and anodes has been for the repair of concrete, the literature is also lacking in life-cycle studies of these technologies when building a new bridge or installing a new bridge deck over an existing superstructure. HPS Evidence of Field Performance HPS has gained general acceptance. Approximately 500 HPS bridges have been constructed in 47 states since the first bridge project in 1997. Bridge owners are specifying HPS to build cost-effective structures with improved strength, weldability, toughness, and corrosion resistance. The HPS-70W grade is classified as a weathering steel that is suitable for use in the un- painted condition. Assessments have been made to investigate the protective rust patina that forms to provide the corrosion resistance of the material. The assessments have included visual inspection and laboratory testing of physical samples. HPS girders that are designed, fabricated, and constructed according to state standards and national specifications have been perform- ing very well (Barth and McConnell 2010; Wilson and Raff 2012; Wiss, Janney, Elstner Associates 2013). Life-Cycle Cost Comparisons An HPS bridge cost comparison was prepared by HDR Engineering and the University of Nebraska–Lincoln, and presented at the HPS Bridge Workshop on October 22, 2007. The study compared weight, girder depth, and cost of 49 girder configurations for a range of span lengths, girder spacing, and steel types. The weight savings for a typical bridge using the HPS 70W instead of the lower strength HPS 50W were in the range of 8 to 14 percent. The cost savings for a typical HPS 70W bridge were in the range of 2 to 9 percent factoring in the slightly higher cost of the HPS 70W material. The hybrid designs were the most economical when using HPS 70W on the most highly stressed plates in the girders and HPS 50W on the other plates (Power et al. 2007). Additional cost savings can be realized when using HPS 70W in shipping and erection, foundations, and reduced approach fill heights. ABC Evidence of Field Performance Although the use of ABC is recent relative to the anticipated design life of bridges, indications to date are that field performance is typically at least

PERFORMANCE OF THE IBRC BRIDGES 43 as good as conventional construction. A state with significant experience with and implementation of ABC is Utah. Although Utah did not have ABC projects funded by the IBRC program, the Utah Department of Transporta- tion (UDOT) made ABC a common practice during the past decade and was the first state to do so. The state has built the largest population of bridges constructed with ABC technologies in the country. The department published an extensive evaluation of the performance of Utah ABC projects through 2016 (UDOT 2016). The investigators visited 44 Utah bridges constructed with ABC components between 2003 and 2012, completed a cursory inspection of each, and determined general performance. ABC tech- nologies evaluated included full-depth and partial-depth precast concrete deck panels, precast abutment elements, and prefabricated superstructure spans installed using self-propelled modular transporters or moved into place with lateral or longitudinal slides. Summary findings were that bridges built with ABC details similar to current UDOT standards were generally performing very well and that bridges that did not adhere to the standards were generally performing fairly. A 2012 report of the National Cooperative Highway Research Pro- gram’s (NCHRP’s) Domestic Scan program, Best Practices Regarding Per- formance of ABC Connections in Bridges Subjected to Multihazard and Extreme Events, described investigations of ABC projects in eight states (Kapur et al. 2012). Findings included that ABC connections were gener- ally perceived to perform the same as conventional connections over time. Life-Cycle Cost Comparisons The states vary in their approaches to evaluating life-cycle costs of ABC relative to conventional construction. Utah, the first state to move to ABC as a standard practice, does not compare costs of ABC with conventional bridge construction. Instead, the state prioritizes traffic mobility and esti- mates project costs based on project limitations. The state also focuses on reducing construction schedules, thereby lessening impacts to the traveling public, and minimizing total project costs. It uses ABC in all projects for which a reduction in total project cost (price plus time) can be achieved. For total project costs, Utah includes both direct construction costs and indirect costs such as maintenance and delay-related user costs. The state evaluates impacts to the public by considering maintenance of traffic, construction schedule, and project-specific critical features such as environmental and railroad constraints. The state uses its own ABC decision-making process (UDOT 2017, 20-3). The Connecticut Department of Transportation (CTDOT) has also developed its own ABC decision process methodology (CTDOT 2017) to assess the viability of ABC technologies during the preliminary design phase

44 PERFORMANCE OF BRIDGES of projects involving the replacement of bridge decks, superstructure spans, or entire bridges. The state’s ABC Decision Matrix has been adopted as a bridge design standard practice in Connecticut. A User Guide explains the use of the ABC Decision Matrix worksheet and definitions of the input variables. The methodology determines the effect of ABC on the overall cost of the bridge, with overall cost including bid price, the cost of managing the project (construction engineering and inspection costs), and road user impacts. Preliminary road user impacts are assessed by estimating and com- paring the road user delay time for conventional construction to a proposed ABC construction methodology. The ABC design methodology is strongly considered when the results of the worksheet analysis are favorable for ABC. Values of other parameters in the matrix may still lead to a favorable ABC rating and possible decision to use ABC in a project regardless of the level of road user impact. CTDOT has implemented this ABC evaluation process in more than 30 projects to date. Table 4-2 shows an example of an application of the construction cost and user delay comparisons that are components of the Connecticut DOT methodology. The example is a comparison of ABC and conventional construction for replacement of a bridge on a state highway. In this case, it was estimated that ABC would have a lower construction cost than the conventional method and would save road users delay by reducing the duration of road closure from 90 days with conventional construction to 49 days with ABC. A set of decision-making tools for use by highway agencies to quanti- tatively determine whether ABC would be beneficial for a specific project, compared with conventional construction, was developed in a 2011 project sponsored by FHWA and eight state departments of transportation through the Transportation Pooled Fund Program (Doolen et al. 2011). The project included development of software and a user manual. Other studies have documented the life-cycle cost advantages that ABC provides over conventional construction. For example, a 2017 case study of the M-100 over CN Railroad Bridge replacement project in Michigan showed that the economic impact of conventional construction on sur- rounding businesses was 16 times greater than the economic impact of ABC (Yavuz et al. 2017). DATA FROM THE STATES ON PERFORMANCE OF THE IBRC BRIDGES The 10 state highway agencies that were interviewed for this study pro- vided information on the performance of bridges in their systems that were the sites of IBRC projects, including records of the most recent results of

PERFORMANCE OF THE IBRC BRIDGES 45 TABLE 4-2 Example of a Comparison of ABC and Conventional Construction Cost Project: Replacement of a bridge over a stream on a state highway Project Alternatives Conventional construction method: Integral bridge with precast abutments, wingwalls and beams; cast-in-place deck and parapets ABC method: Precast rigid frame, footings, and wingwalls User Delay Impact Comparison Average daily traffic (ADT): 4,100 vehicles per day Delay time per vehicle during construction: 20.98 minutes Construction impact duration: Conventional construction 90 days ABC 49 days Aggregate delay: (= ADT × [delay per vehicle] × duration): Conventional construction 5,376 person-days ABC 2,927 person-days User impact change with ABC: –2,449 person-days (negative indicates delay reduction with ABC) Construction Cost Comparison Estimated conventional construction project cost: $2,624,000 Estimated premium for ABC: –10% (negative indicates ABC estimated to be less than conventional construction cost) Construction cost change with ABC: –$262,400 Estimated construction engineering and inspection (CE&I) costs per month: $23,563 Time difference with ABC: –1 month CE&I cost change with ABC: –$23,563 Summary Construction cost change with ABC: –$262,400 CE&I cost change with ABC: – 23,563 Net cost change with ABC: –$285,963 User impact change with ABC: –2,449 person-days SOURCE: Fields and Heredia 2018, 19.

46 PERFORMANCE OF BRIDGES routine inspections of the bridges, as well as observations on performance in response to interview questions. Historical inspection results were analyzed to determine if bridges that were the sites of IBRC projects exhibited accelerated or reduced rates of deterioration, as compared with experience and other analyses of National Bridge Inventory (NBI) data (Nasrollahi and Washer 2015). NBI data from past inspections for the IBRC projects identified for each of the 10 inter- view states were obtained from the FHWA Long-Term Bridge Performance (LTBP) Program bridge portal. These data provide a component-level rating for the deck, superstructure, and substructure of the bridge on a rating scale that varies from 0 (Failed condition) to 9 (Excellent condition). The data include the most recent inspection results and historical inspection records. The period for which historical records were available varied by state and by bridge, from 35 years (1983 to 2017) to 7 years (2011 to 2017). The data provided by the bridge portal were compared with the project descriptions in the FHWA IBRC database and inspection results provided by the interviewed states to identify inconsistencies. There were 121 IBRC projects in the interview states identified in the FHWA IBRC database. From these 121 projects, inspection records for 73 bridges were examined. The data indicate that the deterioration pattern for these 73 bridges is typical of deterioration for all highway bridges. The average condition rating for the deck, superstructure, and substructure components of the bridges was 7.1, 7.3, and 7.1, respectively. A condition rating of 7 is con- sidered “Good” condition. The average condition rating for the deck components included in the study was 7.1, while the lowest condition rating for the deck components was 5 (Fair) for two bridges. However, examination of the historical record for one of these two bridges showed that the rating was unchanged from 1998, indicating that the deck was not part of the IBRC project completed on that bridge. The FHWA IBRC database indicated that the IBRC-funded project consisted of FRP repairs to superstructure elements. The second bridge was an FRP glulam repair, according to the FHWA IBRC database. There were insufficient data available to determine if the IBRC technology was related to the low condition rating, and it was notable that the super- structure and substructure components were rated 5 (Fair) and 4 (Poor), respectively. These data indicate that the bridge overall had a high level of deterioration. The FRP glulam repair was likely applied to a deteriorated bridge to extend its service life, and therefore these data do not reflect the performance of the IBRC technology. Several IBRC projects involved the simultaneous construction of a bridge deck using corrosion-resistant reinforcing paired with the construc- tion of a bridge deck using conventional reinforcing. The two bridge decks were located on adjacent bridges, such that the structures were exposed to

PERFORMANCE OF THE IBRC BRIDGES 47 similar traffic levels and environmental conditions. For example, Missouri project MO 2000-01 included the construction of two decks on adjacent bridges. One deck was constructed using solid stainless steel rebar and the second was constructed using epoxy-coated rebar. The current condition rating for each bridge is 7 (Good). New Hampshire project NH-2002-01 featured a pair of bridge deck replacements for bridges carrying I-93 over a railway. One deck was constructed with low-chromium steel rebar and the second was constructed with epoxy-coated rebar. Both decks are cur- rently rated an 8 (Very Good). A second IBRC project in New Hampshire (NH-2002-02) consisted of twin bridge deck replacements with galvanized rebar in one deck and epoxy-coated rebar in the second deck. Both decks have current ratings of 8 (Very Good). These data illustrate that the IBRC technologies are performing in a similar manner to conventional technolo- gies. It is not possible to predict future performance based on the available information. To evaluate the overall performance of bridge decks constructed with IBRC technologies, the condition ratings were analyzed for 43 projects that included renovation or construction of decks. The historical condition rat- ings for the 43 decks starting from 2005, when the IBRC program ended, through the most recent available inspection year (2017) were tabulated and the average condition rating versus years elapsed since IBRC project construction was calculated. For comparison, average deck condition rat- ings were tabulated for a sample of 10 bridges in each of the same states that are less than 150 ft in length and were constructed in 1999. As shown in Figure 4-1, the average condition rating of the IBRC bridges was slightly over 8 (Very Good), 6 years after the IBRC project, and had diminished to 7 (Good) by 17 years after the project. In the figure, deterioration rates of the IBRC and non-IBRC bridges appear closely similar. A 17-year period is too short, compared with the intended service lives of the decks, to fully judge comparative performance of alternative technologies. However, the graph indicates that the available bridge rating data provide no indication of inferior or superior performance for the IBRC decks, compared with conventional construction. The average condition rating for the superstructure, for the 73 IBRC bridges in the interview states for which inspection records were examined, was 7.3. Six projects had a superstructure condition rating of 5 (Fair). These included the previously identified glulam project and two FRP deck projects in which the IBRC technology was not implemented on the super- structure component. Therefore, the relatively low superstructure condition rating for these three bridges cannot be ascribed to the IBRC technology. Two of the bridges with superstructure condition rating 5 (Fair) were HPC projects involving bulb tee girders in Virginia. According to the inter- view with Virginia Department of Transportation staff, construction- and

48 PERFORMANCE OF BRIDGES design-related problems with the bulb tee girders resulted in the relatively low condition rating for the two bridges. The design of the bulb tee gird- ers included a narrow web with inadequate space for longitudinal tendon ducts. The inadequate space resulted in longitudinal cracking and spalling in the girder webs at the time of construction (Sprinkel and Balakumaran 2017). The state is no longer using the design as a result of the experience from the IBRC program. Finally, one project on a bridge with a present superstructure condi- tion rating of 5 (Fair) was an externally bonded FRP reinforcement project to repair a superstructure that was deteriorated at the time of the project. According to the historical inspection records, the superstructure compo- nent was also rated 5 at the time of the IBRC program, and the rating was increased to 6 during the time interval of 2006–2011. In other words, the externally bonded FRP reinforcement was used as a repair that improved the condition of the member. In summary, the inspection results for the superstructure component did not show any unusually rapid deterioration of bridges that appeared to be related to IBRC technologies or the IBRC program. FIGURE 4-1 Average condition rating for bridge decks in IBRC bridge deck projects in interview states, compared with condition of decks in a sample of non-IBRC bridges in the states constructed in 1999. SOURCES: State highway agency bridge inspection reports, FHWA NBI database. 5 6 7 8 9 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Av er ag e Co nd iti on R at in g Years Since Construction IBRC Bridges Deck Non IBRC Bridges Deck Excellent Very Good Good Satisfactory

PERFORMANCE OF THE IBRC BRIDGES 49 The substructure component had an average rating of 7.1. The lowest component ratings for the substructure were two bridges with a rating of 4 (Poor). These were the previously mentioned glulam project and one other bridge in which the IBRC technology was an FRP deck and therefore the substructure condition is not related to any IBRC technology. There were four projects with substructure components with a rating of 5 (Fair), two of which were externally bonded FRP reinforcement repair projects and two of which were projects involving superstructure elements. There was no evidence found that IBRC technologies involved in these projects were subject to accelerated deterioration. Data were analyzed to determine the overall performance of corrosion- control technologies for reinforcing steel. This analysis considered solid and clad stainless steel rebar and low-chromium steel rebar used in the construc- tion of decks. Ten projects were identified. The average deck condition rat- ing was 7.2, with three decks rated an 8 (Very Good), six rated 7 (Good), and a single deck rated as a 6 (Satisfactory). The data were also analyzed to determine the current condition of elements formed from HPC. There were 16 projects that used HPC. The average ratings for the superstructure, substructure, and deck were 7, 7.25, and 7.31, respectively. Overall, the analysis indicated that based on the available data, bridges that were part of the IBRC program had typical deterioration patterns. It should be noted that the IBRC program was completed in the time period of 1998–2005, and therefore components constructed or repaired during the program have been subjected to 13 to 20 years of service, which is a relatively short period of time for a bridge to deteriorate significantly, con- sidering the minimum 75-year design life of bridges built today. Therefore, it is difficult to determine at this time if the use of the IBRC technologies is effectively extending the service life of bridges as compared with conven- tional technologies.