Current Practices and Guidelines for the Reuse of Bridge Foundations (2017)

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.

7 chapter two Literature review Reuse of bridge foundations involves concepts from many disciplines. Accordingly, a relatively broad review of literature is necessary to appreciate the benefits and challenges of foundation reuse. Many of the concepts discussed in this chapter are also defined in the Glossary. Nine topics of interest to reusing bridge foundations were reviewed and provide the outline for this chapter: 1. An overview of U.S. bridges near the end of their service life focusing on agency practices for managing and inspecting bridge inventories; 2. History of foundation reuse, including international and U.S. experience; 3. Challenges for foundation reuse identified by early U.S. experience; 4. Investigation and condition assessment methods for existing foundations; 5. Deterioration mechanisms affecting bridge foundations, including deterioration of steel and concrete, as well as scour; 6. Methods for predicting remaining service life for foundations; 7. Techniques to improve load capacity of existing foundations; 8. Decision methods related to reuse of foundations, and 9. Notable examples of foundation reuse. u.S. BridgeS Near the eNd of Service Life: overview aNd ageNcy PracticeS FHWA compiles the National Bridge Inventory (NBI), a database of all U.S. bridges greater than 20 feet in length with roadways passing above or below. In 2015, there were more than 610,000 bridges in the NBI (FHWA 2015a). The following is a brief introduction to U.S. policies regarding bridge inspections, the state of U.S. bridges as revealed by those inspections, and prac- tices used by U.S. agencies to manage bridge inventories. Bridge inspection information contributes to foundation reuse decisions, which agencies frequently consider within the framework of bridge management. The information in this section also provides the context for the economic motivation for foundation reuse. Bridge inspection and condition ratings All NBI bridges must be inspected every 2 years according to National Bridge Inspection Stan- dards. The Standards assign a condition rating of from 0 to 9 for the bridge deck, superstructure, and substructure. The Standards do not address inspection of bridge foundations explicitly; how- ever, limited information regarding foundation conditions may be included with the substructure inspection information. Descriptions for each rating value are shown in Table 1 (FHWA 1995). NBI applies the term “structurally deficient” to bridges with ratings of 4 or less; the designation does not necessarily imply a lack of safety, but often requires reduced load ratings or speed limits. NBI applies the term “functionally obsolete” to bridges that do not satisfy standards relating to roadway safety or mobility. In 2015, there were 58,791 bridges deemed structurally deficient and 84,124 deemed functionally obsolete, representing 10% and 14% of all NBI bridges, respectively (FHWA 2015a).

8 Bridge Management U.S. transportation agencies use the NBI ratings to inform decisions regarding allocation of agency resources to maintain the inventory of agency bridges. The FHWA Bridge Preservation Guide (2011) defines three general actions taken by agencies as part of bridge management: • Preventive maintenance is a planned strategy of cost-effective treatments that preserves the system, retards future deterioration, and maintains or improves the functional condition. • Rehabilitation is major work required to restore the structural integrity of a bridge. • Replacement refers to total replacement of a structurally deficient or functionally obsolete bridge with a new facility constructed in the same general traffic corridor. Brief hiStory of fouNdatioN reuSe Foundation reuse is not a new concept, with international experience dating back centuries. This sec- tion introduces the history of foundation reuse abroad and in the United States, including information about recent initiatives promoting foundation reuse. international CIRIA, a United Kingdom-based construction trade group, published Reuse of Foundations in 2007 (Chapman et al.). The report begins with a section on “historical context,” which explains that foun- dation reuse in Europe “used to be the norm rather than the exception.” In late 16th century London, new buildings were only allowed if they were placed on old foundations, a rule aimed at preventing urban sprawl. The report attributes a modern decline in foundation reuse to heavier structures and higher performance expectations; for example, less tolerance for architectural cracking. However, the CIRIA report points to recent trends that have renewed applications of foundation reuse: urban underground space congested with utilities, transit tunnels, and existing basement walls; a desire to preserve archaeological remains; and the high load capacity of some existing foundations supporting structures slated for replacement. In 2003, the European Union initiated a research project to investigate foundation reuse. The proj- ect was titled “Re-use of Foundations for Urban Sites” (RuFUS), and its partners included construc- tion trade groups, engineering organizations, and engineering consulting firms from across Europe. In 2006, the project partners published Reuse of Foundations for Urban Sites: A Best Practice Hand- book (Butcher et al. 2006). Although the RuFUS manual is geared toward European practice and Rating Description of Condition 9 Excellent 8 Very good: No problems are noted. 7 Good: Some minor problems are noted. 6 Satisfactory: Structural elements show some minor deterioration. 5 Fair: All primary structural elements are sound, but may have minor section loss, cracking, spalling, or scour. 4 Poor: Advanced section loss, deterioration, spalling or scour are noted. 3 Serious: Loss-of-section deterioration of primary structural elements is noted. Fatigue cracks in steel or shear cracks in concrete may be present. 2 Critical: Advanced deterioration of primary structural elements is noted. Fatigue cracks in steel or shear cracks in concrete may be present or scour may have removed substructure support. Unless closely monitored it may be necessary to close the bridge until corrective action is taken. 1 Imminent failure: Major deterioration or section loss is present in critical structural components or obvious vertical or horizontal movement is affecting structural stability. Bridge is closed to traffic, but corrective action may put it back in light service. 0 Failed: Out of service and beyond corrective action. Source: FHWA (1995). TABLE 1 NBI CoNDITIoN RATINGS FoR BRIDGE DECK, SUpERSTRUCTURE, AND SUBSTRUCTURE

9 building foundations, its review of and recommendations regarding the technical and administrative challenges of foundation reuse are also relevant to bridge foundations and to U.S. practice. The RuFUS manual is a useful supplement for much of the information presented in this chapter. united States The history of foundation reuse in the United States is not well documented. Widespread attention to bridge foundation reuse among U.S. transportation agencies has developed only recently and largely as an outgrowth of work focused on bridges with “unknown foundations.” The term unknown founda- tions is applied within NBI to designate bridges over waterways for which hydraulic risks cannot be assessed because of missing bridge plans. In 2013, FHWA created the Foundation Characterization program to address the issue of unknown foundations. The program considers not only hydraulic risks but also unknown foundation issues pertaining to changes in service loads, geo-hazards, and reuse of foundations. The program includes foundations for all bridges, not just those over waterways. The Foundation Characterization program includes in its characterization mission information regard- ing foundation type, embedment depth, geometry, material, foundation integrity, and load capacity (Schaefer and Jalinoos 2013). These characteristics are sought for all foundations, even those not designated in NBI as unknown. In 2013, North Carolina DoT polled 18 state transportation agencies regarding their experi- ence with foundation reuse. The results were reported by Collin and Jalinoos (2014). Ten agencies responded to the survey request. Seven had reused an existing foundation for bridge replacement, eight had retrofitted an existing foundation, and all ten had done one or the other. Although all of the agencies had experience with foundation reuse, none of the surveyed agencies reported having policies or guidelines regarding foundation reuse. chaLLeNgeS of fouNdatioN reuSe In 2013, FHWA’s Foundation Characterization program hosted a stakeholder workshop to discuss program goals. participants included FHWA, representatives from five state transportation agencies, academia, and industry. The participants identified four major challenges for reuse of existing bridge foundations: 1. Condition assessment: What is the structural integrity of the existing foundation? 2. Load capacity: How can ultimate resistance of the existing foundation be determined, and is it greater than the value from the original design? 3. Remaining service life: How much longer can the existing foundation be expected to maintain serviceability? 4. Design codes: How should existing foundations be considered within the context of estab- lished design codes and specifications, which were developed for new foundations? The four challenges are closely related to the topics presented in the rest of this literature review and to the concepts discussed throughout the rest of this Synthesis. Collin and Jalinoos (2014) reported on a workshop specific to foundation reuse held at the 93rd Annual Meeting of TRB in 2014. The attendees developed another list of challenges. Some items in the list overlap with the 2013 Foundation Characterization program list and most are more specific than the Foundation Characterization program list. Some of the challenges are listed here: • New structures often have greater foundation loads. • Critical documentation for foundations is often unavailable. • Designers are not confident information in historical records is accurate. • Construction and monitoring standards at time of original construction were different from those in place today. • There is a lack of existing state and/or federal guidance.

10 one of the presentations from the TRB workshop introduced questions related to risk and respon- sibility arising in projects involving foundation reuse (Brown 2014). The presentation focused on allocation of risk among transportation agencies, consulting engineers, and construction contractors. The presentation suggested questions surrounding the reliability of historical records, performance risks associated with reuse, and that investigation of existing foundations create an uncertain stan- dard of care for engineers charged with designing foundation reuse projects. iNveStigatioN aNd coNditioN aSSeSSMeNt of exiStiNg fouNdatioNS Ideally, consideration of an existing foundation system for reuse would be based on reliable informa- tion regarding the foundation elements, including • Type of foundation, including material type; • Dimensions of the foundation elements; • Location of the foundation elements; • Depth of foundation; and • Structural integrity of the foundation. The first four types of information are referred to in this Synthesis as the “primary details” of the existing foundation. Evaluation of the last type of information listed, structural integrity, is com- monly referred to as “condition assessment.” Information regarding the load capacity of a founda- tion is also important to reuse evaluations and is considered in its own section, which introduces various methods of collecting information regarding existing foundations. The first method, review of historical records, is often used to define the scope of the remaining methods. Conventional field investigations (excavation and probing, concrete coring, and laboratory testing) are presented before geophysical and nondestructive test methods. presentation of geophysical methods is organized by type: pile integrity test methods, surface methods, and borehole methods. The information regarding geophysical methods presented in this Synthesis is introductory and only addresses eight of the many available methods. Wightman et al. (2004) presented more detailed and background information in FHWA’s Application of Geophysical Methods to Highway Related Problems. historical records Historical records include all agency documentation that might contain information regarding pri- mary details or structural integrity of the bridge foundation. • plan drawing sets from the original design. • Specifications or special provisions from the original construction. • Construction records, including – plan drawing sets showing final, as-built dimensions and details; that is, as-built plans or drawings, and – Foundation installation records (e.g., pile driving logs and load tests). • Inspection reports. • Monitoring reports (e.g., for scour-critical bridges). • Documentation of any extreme events impacting the bridge (e.g., flooding and barge impact) or seismic loading. primary details are typically included in plan drawings and/or construction records; however, plans and construction records are not always available. Historical records likely contain little if any information regarding the structural integrity of a foundation, although the information in inspec- tion and monitoring reports may occasionally be used to infer some limited information regarding foundation performance. Historical records are typically gathered before any field explorations are completed. The reliabil- ity of historical records is not certain; evaluating the accuracy of historical information is a primary

11 goal and challenge of existing foundation investigations. The scope of field explorations is generally intended to verify and supplement the information from historical records. The remaining investiga- tion methods in the following sections are all field exploration methods. excavation and Probing Excavation to reveal all or portions of a foundation is the most direct method of determining infor- mation regarding an existing foundation. An example of an excavation test pit is shown in Figure 1. The depth of excavations is typically limited owing to safety, adjacent roadways, groundwater, and cost; for many projects, these constraints may prohibit excavations altogether. Some factors associ- ated with deterioration, including unsaturated soils and groundwater fluctuation, are often near the surface. In addition, most agencies are equipped to excavate test pits inhouse. probing generally refers to inserting a small diameter rod into the ground to identify the depth of the top of a footing or pile cap. probing is quick and inexpensive; however, the presence of cobbles, boulders, or random fill can result in false positives (olson et al. 1998). concrete core drilling and Laboratory testing Shallow foundations and pile caps can be investigated by drilling to the top of the footing or cap and then coring through the foundation concrete. Coring can also be completed through drilled shaft and concrete pile foundations. Visual inspection of the cores provides some indication of concrete quality (e.g., observation of concrete cracking), and laboratory tests of core samples can be used to evaluate concrete compressive strength and potential durability concerns (e.g., tests of chloride penetration to evaluation potential for corrosion of reinforcing steel; see “Deterioration of Concrete” in this chapter). In addition, FHWA’s Foundation Characterization program has demonstrated the use of geophysical logging technology through core holes for detailed characterization of concrete durability and structural integrity (Jalinoos 2015; Jalinoos et al. 2016). Geophysical methods are discussed here. overview of geophysical and Nondestructive test Methods Geophysical and nondestructive test methods include pile integrity test methods, surface methods, and borehole methods. pile integrity test methods are applied at the top of foundation elements to assess structural integrity and estimate length (depth). Surface geophysical methods are those that can be implemented with equipment at the ground surface. Unlike the pile integrity methods, the surface geophysical methods used to evaluate existing foundations generally involve interpreting FIGURE 1 Test pit exposing timber piling (Source: Johnson and Chauvin 2013).

12 measurements of signals that have passed through the ground and the foundation, whereas pile integ- rity test measurements only involved signals from within the foundation element. Borehole methods require drilling boreholes close to the existing foundation. The borehole requirement makes this method costlier than surface methods, and for some bridges borehole methods may be impractical or impossible. However, these methods are generally more successful and reliable than surface meth- ods for identifying the depth of deep foundation elements. Specific geophysical and nondestructive test methods commonly applied to investigate existing foundations are introduced in the following sections. Pile integrity test Methods Sonic Echo/Impulse Response (SE/IR) methods involve striking the surface of a foundation ele- ment and measuring the reflected compression wave at a receiver installed on the foundation surface. Schematics of various test configurations are shown in Figure 2. Interpretations of the reflection time (SE) or frequency (IR) can indicate the approximate location of the bottom of the foundation or be used to identify defects such as soil inclusions. The test can be done quickly and at relatively low cost; however, the information provided is often limited. It cannot be used with H-piles or for foundations embedded in rock, and its depth range is limited to about 25 feet when foundations are installed in very stiff soils (Wightman et al. 2004). The Bending Wave method is quite similar to the SE/IR tests; however, bending waves are measured rather than compression waves. The method was developed by the North Carolina DoT to evaluate the length of timber piles supporting scour-susceptible bridges (Holt et al. 1994). The method is applied in a manner similar to SE/IR tests, but the foundation element is struck on its side in order to generate bending waves. The Bending Wave method has all the limitations of the SE/IR test; in addition, attenuation issues associated with bending waves may result in further restrictions on the maximum depth that can be interpreted. The Ultraseismic method is similar to the SE/IR and Bending Wave methods, but triaxial receiv- ers are used to interpret measurements of three wave modes rather than just considering compression or bending waves. A typical test configuration is shown in Figure 3. The test requires that at least 5 feet of the foundation be exposed. The method was developed to improve on SE/IR and Bending Wave tests, especially for structures such as bridges with many reflecting boundaries (Wightman et al. 2004). The method is generally more reliable than the SE/IR or Bending Wave methods. FIGURE 2 Application of SE/IR methods for various top-of-foundation configura- tions. Configuration (a) produces the best results; configuration (c) is the most difficult to interpret (Source: Wightman et al. 2004).

13 Surface geophysical Methods Seismic methods include Seismic Reflection, Seismic Refraction, Spectral Analysis of Sur- face Waves, Multichannel Analysis of Surface Waves, and tomography. The methods generally involve interpreting features such as soil layer boundaries based on stiffness contrasts that influ- ence seismic wave velocities. Applied to investigations of existing foundations, the methods are intended to identify foundation depth from stiffness contrasts between the foundation material and the soil that surrounds it. The success of seismic methods for unknown foundation applica- tions has generally been limited by the source energy required to identify foundation depths with useful precision. However, a recent application of Full Waveform Tomography by Nguyen et al. (2016) produced promising results. Nguyen et al. installed two drilled shafts 4 feet (1.2 m) in diameter and 50 feet (15 m) long at an experimental site owned by Florida DoT. The researchers installed a linear array of 23 geophones adjacent to the shafts and recorded 31 shots from a 20-lb hammer, and then per- formed full waveform inversion to shear and compression wave velocities as shown in Figure 4. FIGURE 3 Ultraseismic method test configuration (Source: Wightman et al. 2004). Distance (m) D ep th (m ) FIGURE 4 Shear wave velocity contours from full waveform tomography. Contours accurately indicate known shaft depths of 50 ft (15 m) (Source: Nguyen et al. 2016).

14 The contours clearly indicate the shaft locations and accurately identify shaft depth of up to 50 feet. That the geophones were installed next to the shafts, not in line with the shafts, is noteworthy since placement in line with the shafts would not be feasible for many bridges. Electrical methods are used to identify materials based on variations in their electrical properties. The methods are feasible because electrical resistivity (ER) values for different geologic materials vary considerably (Wightman et al. 2004). The methods generally involve injecting electrical cur- rent into the ground at two sources and then measuring electrical potential difference at two different points between the sources. A recent example of electrical methods applied to an unknown foundation is provided by Tucker et al. (2015). The researchers used ER and induced polarization (Ip) methods to examine two bridge sites, one with a known and one with an unknown foundation. The results from the unknown foundation site are shown in Figure 5. Analysis by Tucker et al. of the Ip data revealed that the bottom of the pile is 27-feet (8.2-m) deep, consistent with the light contours and near actual depth of 32 feet (9.8 m). Tucker et al. concluded that ER was more effective near the ground surface; however, Ip was more successful at identifying foundation depths. Electrical methods also include field corrosion probes to evaluate corrosion of steel piling, which is discussed further in the “Deterioration of Steel” section. NCHRP Report 408 (Beavers and Durr 1998) documents development of field corrosion probes to measure galvanic currents, polarization resistance, and corrosion potential of steel in soil environments. The corrosion probe consists of sev- eral electrically isolated rings to measure galvanic current. The rings can be electrically coupled to simulate steel piling or left isolated to evaluate specific depths. The procedure developed by Beavers and Durr recommends installation of the probes near the actual (or design) pile location and through any soil layers identified as having moderate or high corrosion potential. procedures associated with using field corrosion probes to measure galvanic current and polarization resistance were incorpo- rated in AASHTo Standard R 27-01 (2015). Ground penetrating radar (GPR) is an electromagnetic method that detects dielectric contrasts to locate interfaces between geologic strata, the groundwater surface, or buried structures such as tanks and utility pipes. There are various forms of GpR, with the primary difference being the characteristics of the radar antenna. Higher frequency antennas are associated with greater resolution, whereas lower frequency antennas can detect features at greater depth. For investiga- tion of existing foundations, GpR is limited to identifying shallow foundation dimensions and the reliability of such measurements may be low (olson et al. 1998). GpR can also be used to assess the near-surface integrity of concrete and timber structures (Wightman et al. 2004). FIGURE 5 (a) Inverted ER cross section and (b) inverted IP cross section for unknown foundation bridge site (Source: Tucker et al. 2015).

15 Borehole geophysical Methods The Parallel Seismic method applies the same principles used in the pile integrity test methods; however, the collection of data in a borehole adjacent to the existing foundation addresses many of the limitations regarding depth and the effect of pile caps or other structures at the surface. As shown in Figure 6, the method involves striking the existing foundation and recording seismic waves in an adjacent borehole. The energy source can be applied to any exposed surface on the foundation or on the structure connected to the foundation (e.g., a pile cap). The borehole should be as close to the foundation as possible and generally must be within 5 feet of the foundation (Wightman et al. 2004). The borehole should also extend 10 to 15 feet below the bottom of the existing foundation. Results of the parallel Seismic method generally are more reliable than the pile integrity and surface-based methods; foundation depth measurements are generally 95% accurate or better (Wightman et al. 2004). The primary limi- tations of the method are the cost and physical constraints associated with the borehole. Cross-Borehole Tomography is used to generate images of the subsurface between boreholes based on travel times of seismic waves traveling between the boreholes. Typically, the source is acti- vated at closely spaced depths in one borehole and measurements are collected from closely spaced receivers in the other borehole. Example results from a Cross-Borehole Tomography test at a site with steel piles driven to sandstone are shown in Figure 7. Interpretation of the test results correctly identified the pile depths (Kase et al. 2003). deterioratioN MechaNiSMS affectiNg Bridge fouNdatioNS Deterioration—the loss of structural integrity over time—is a primary concern in foundation reuse applications. This section presents an introduction to how foundation materials deteriorate and directs the reader to additional resources. The section also includes an introduction to scour, another mechanism that can threaten foundation integrity over time. deterioration of Steel Steel foundations primarily consist of H-piles and pipe piles, which are frequently backfilled with concrete. The primary mechanism for deterioration of steel is corrosion, which is an electrochemical FIGURE 6 Configuration of Parallel Seismic test method (Source: Wightman et al. 2004).

16 process wherein air, water, and negatively charged ions oxidize steel, resulting in its degradation. In addition, corrosion of steel reinforcement is a primary concern regarding concrete durability, as discussed in the following section. This section discusses subsurface conditions associated with cor- rosion of steel piling and summarizes three examples where steel piling has corroded severely. In all three examples, the piles were installed in aggressive subsurface environments. Below the groundwater table soil is generally free of oxygen. As a result, most ground conditions inherently protect steel piling from corrosion. However, zones of groundwater fluctuation and fill materials above the groundwater table are associated with moderate corrosion. Certain aggressive ground conditions are associated with severe corrosion. FHWA’s Design and Construction of Driven Pile Foundations manual (Hannigan et al. 2006) identifies marine environments and sites with contaminated soil and groundwater as aggressive. NCHRp Study 10-46 investigated corrosion of steel piles in nonmarine environments. The results are documented in NCHRP Report 408 (Beavers and Durr 1998). The study found that aggressive sites associated with contaminated soil and groundwater can be identified by tests of soil samples to reveal high chlorides, high sulfate ion concentration, or low soil resistivity. Beavers and Durr concluded that the mechanisms for underground corrosion are understood, but attempts to predict specific values of soil corrosivity “have met with limited success.” AASHTo Standard R 27-01 (2015) was developed out of NCHRp project 10-46. The standard is an assessment procedure for corrosion of steel piling in nonmarine environments that establishes rela- tive corrosion risks and recommended testing procedures based on the risk level. The standard also includes guidance for the reuse of steel piling, including methods for predicting the remaining service life from observations of pile condition and rate of deterioration. The Leo Frigo Memorial Bridge recently drew considerable attention to the issue of corrosion of steel piling in aggressive environments. The bridge carries I-43 over the Fox River at an industrial site in Green Bay, Wisconsin. In 2013, the bridge was closed for emergency repairs after one of the piers settled 2 feet overnight. Subsequent investigation revealed that corrosion of the steel H-piles caused the settlement. Corrosion was attributed to the “highly corrosive combination of industrial porous fly ash, high concentrations of chlorides, high sulfate concentrations, low soil resistivity, and microbial activity” (Wisconsin DoT 2015). The closure brought renewed attention to the risk of corrosion of steel piling. FIGURE 7 Tomograms with interpreted pile depths (Source: Kase et al. 2003).

17 The Leo Frigo case was not the first documented example of severely corroded steel piles. In 1988, Connecticut DoT began replacing the intersection of I-84 and I-91 in Hartford, intending to reuse the existing 30-year-old steel H-piles. Excavation beneath the existing pile cap during construction exposed the tops of the piles, which were observed to be severely corroded. Measurements indicated section loss between 35% and 65%, averaging 55% (Long 1992). The corroded lengths of the piles were in contact with a clay fill pocket between layers of incineration byproduct fill consisting of ash, slag, and cinders. Gu et al. (2015) documented corrosion of steel H-piles for a bridge carrying Girard Avenue and I-95 interchange traffic in philadelphia. prior to the rehabilitation of the 34-year old bridge for a major I-95 improvement project, an investigation of the piles was completed since the piles were known to be installed through cinder ash fill above the groundwater table. The project team extracted one of the existing piles and observed modest corrosion in two locations: (1) just below the pile cap and (2) along a section of pile that had been 14 to 17 feet below the ground surface. Gu et al. speculate that exposure to deicing agent caused the upper corrosion, whereas the deeper corrosion was consistent with the location of the cinder ash fill. The maximum measured section loss was 12%. Design of the retrofitted foundation was completed by reducing the capacity of the existing piles assuming the maximum observed rate of corrosion (i.e., 12% in 34 years) would continue for the 50-year service life of the retrofitted bridge. New piles were included in the design to achieve the founda- tion load capacity requirements for the retrofitted structure. Design of the new piles assumed the same rate of corrosion. deterioration of concrete Many foundation elements are constructed of concrete, including all shallow foundations and drilled shafts and some driven piles, notably precast concrete piles. To consider concrete deterioration, it is necessary to first define concrete permeability. After cement hydration, excess mix water is typi- cally lost to the environment, leaving air voids within the hardened concrete that can transmit fluids. Concrete permeability is its ability to transmit fluids. permeability is an important concrete topic because most concrete deterioration mechanisms are initiated by penetration of water into concrete. Thus, permeability is a useful surrogate for concrete durability, which can be defined as resistant to deterioration. Accordingly, Mehta and Monteiro (2014) developed a holistic model for concrete deterioration that links concrete permeability with durability. The model considers a reinforced concrete structure that contains voids and microcracks. With time, the concrete structure is subjected to two stages of environmental action that increase permeability and promote deterioration. During Stage I, the structure is subjected to applied loads, thermal cycles, and wetting and drying cycles, all of which advance microcracking and increase permeability, although without any apparent loss of service. Eventually, the increase in permeability is sufficient to allow penetration of water and air, both of which potentially subject the structure to a variety of chemical and physical attacks. These attacks mark the initiation of Stage II. The attacks decrease serviceability and increase permeability, result- ing in a cycle whereby Stage II deterioration feeds on itself to rapidly accelerate deterioration. The model forecast may be gloomy, but Mehta and Monteiro concluded that the “no-damage” period corresponding to Stage I can last “hundreds of years by using concrete mixtures that are impermeable and will remain crack-free during the service.” Certain environments can greatly increase the rate at which the general deterioration mechanism described by Mehta and Monteiro occurs. Fortunately, the ground provides excellent protection from many environmental threats; however, several of the threats are nevertheless relevant to foundations: • Alkali–silica reactivity: Some aggregates are siliceous, which makes them acidic, leading to potential reactions with concrete pore fluid, which is alkaline. The reactions produce a gel that expands in the presence of water, leading to concrete cracking. • Freeze-thaw cycles cause water in concrete pore space to expand and contract, which can lead to cracking. • Sulfate attack refers to chemical reactions between cement paste and sulfate ions. The reac- tions produce chemicals that swell in the presence of water, potentially resulting in cracking.

18 Although these mechanisms are relevant to foundation concrete, it is important to note that docu- mented cases of such threats affecting underground concrete are relatively limited. Mehta and Monteiro (2014) is a useful reference for additional information regarding concrete and concrete deterioration. deterioration of timber Timber piling is subject to decay as well as attack by insects (e.g., termites or beetles on land and marine borers in marine environments). FHWA’s driven pile manual (Hannigan et al. 2005) states that timber piling at depths permanently below the groundwater table is generally not subject to decay or insect attack, but piling in zones of water table fluctuation resulting in wet-dry cycles and piling above the water table are subject to decay and attack. Deterioration by decay and attack is likely severe unless the pile is treated with a wood preservative. The FHWA manual states that “fully embedded, treated foundation piles partially above the ground water with a concrete cap will typi- cally last on the order of 100 years or longer.” Scour Scour occurs when water erodes the earthen material surrounding bridge foundations. Scour is the most common cause of bridge failures, affecting both bridge piers and bridge abutments, with most examples of extreme scour occurring in response to flood events. Most U.S. transportation agencies reference FHWA’s Hydraulic Engineering Circular No. 18: Evaluating Scour at Bridges (HEC-18; Arneson et al. 2012) for analysis and design related to scour. In addition to addressing design of new bridges for scour resistance, the manual also addresses prediction of scour for existing bridges and inspection of bridges for scour. The latter two topics are of particular relevance to reusing founda- tions because scour issues are frequently critical to reuse decisions for bridges over water. As shown in Figure 8, FHWA recommends that engineers use HEC-18 alongside two accom- panying circulars: (1) HEC-20, Stream Stability at Highway Structures (Lagasse et al. 2012) and (2) HEC-23, Bridge Scour and Stream Instability Countermeasures (Lagasse et al. 2009). For existing bridges, National Bridge Inspection Standards require scour evaluation for bridges over water as well as development of scour plans of action for bridges deemed scour critical. Scour plans of action frequently include implementation of scour countermeasures to improve scour resistance. HEC-23 includes a matrix of available countermeasures organized by form, including 37 hydraulic countermeasures, nine structural countermeasures, five biotechnical countermeasures, and eight monitoring countermeasures. The matrix indicates how well-suited each countermeasure is for various applications and river environments. reMaiNiNg Service Life Remaining service life was identified by FHWA’s Foundation Characterization program as a primary challenge for reuse of bridge foundations. This section introduces the concept of foundation service life before discussing methods for prolonging the remaining service life of existing foundations. foundation Service Life Designing bridges for service life is a topic of current research and policy interest for many of the same reasons foundation reuse has gained attention: reducing costs and risks associated with aging bridge inventories. A pair of recent Strategic Highway Research program 2 (SHRp 2) research proj- ects have investigated service life: SHRp 2 project R19A, Bridges for Service Life beyond 100 Years: Innovative Systems, Subsystems, and Components (Azizinamini et al. 2013) and project R19B, Bridges for Service Life Beyond 100 Years: Service Limit State Design (Modjeski and Masters, Inc. et al. 2015). project R19A defined important terms related to service life: • “Service Life—The time period during which the bridge element, component, subsystem, or system provides the desired level of performance or functionality, with any required level of repair and maintenance.

19 • Target Design Service Life—The time period during which the bridge element, component, subsystem, or system is expected to provide the desired function with a specified level of main- tenance 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 according to the current version of AASHTo LRFD Bridge Design Specifications” [LRFD (Load and Resistance Factor Design)]. By contrast, the AASHTo LRFD Bridge Design Specifications (2014) definition of service life— “the period of time that the bridge is expected to be in operation”—is less specific and makes no dis- tinction between actual service life and target design service life. Notably, the 75-year design life in the AASHTo specifications is not a target service life but rather a basis for the transient load determination. project R19A also provides a 12-step procedure for designing bridges for service life. The procedure emphasizes the component nature of bridge service life and the location-dependent nature of deteriora- tion. The procedure recommends two general solutions to designs for service life: (1) use of durable materials and (2) implementation of intervention solutions (i.e., repair or replacement of components) that have been optimized using life-cycle cost analysis. Although SHRp 2 project R19A focused on the general design approach for service life consider- ations, project R19B provided more specific guidance, including new load and resistance factors recommended for implementation in AASHTo’s LRFD Bridge Design Specifications. The reliability- based factors were calibrated for several service limit states, including foundation deformations. Implementation of the recommended values of load factor for foundation deformations, gSE, will likely affect the structural design of many bridges since current specifications use gSE = 1, effectively neglecting foundation deformations for structural service life design. FIGURE 8 FHWA’s recommended procedure for analyzing bridge scour. “CM” refers to scour countermeasure (Source: Arneson et al. 2012).

20 The foundations deformation work of project R19B addresses a critical serviceability consideration in designing for service life; however, methods related to the deterioration component of service life are lacking. The same could be said for many bridge components, but is especially true for founda- tion elements, primarily because of the inherent condition assessment challenges. FHWA is currently researching remaining service life for foundations, including performance measurements for existing foundations, the effect of service environment on the remaining service life of foundations, and the effect of foundations and other geotechnical features on overall system performance (Nichols 2016). techniques to improve remaining Service Life Methods of prolonging the service life of a foundation system involve retrofitting individual founda- tion elements to address deterioration or replacing deteriorated elements outright. Figure 9 is pile repair decision matrix created by Wan et al. (2013) for Wisconsin DoT as part of a research effort that developed a substructure repair guide. The matrix includes multiple repair options for timber, steel, and concrete piles. Cost estimates and repair service life values are also listed for methods where data were available. The substructure repair manual proposed by Wan et al. includes addi- tional information regarding each method. Most of the methods involve some form of encasement or sleeve around deteriorated lengths of pile. Examples in the report typically show repairs imple- mented near the water surface in piles extending above the mudline (for bridges crossing water); however, it is feasible the repairs could be implemented at shallow depths. Agencies frequently implement repairs to prolong the service life of the substructure above the foundation, which is subject to more deterioration than the foundation by virtue of being above ground. Concrete patching and other surface repairs are routinely incorporated into preservation efforts. Many of the foundation reuse examples involving bridge replacement include substructure concrete surface repairs, as well as more intensive efforts to halt and prevent corrosion of concrete reinforcement. one such measure commonly encountered in bridge foundation reuse literature is electrochemical chloride extraction (ECE), which is documented here. The literature review did not encounter any examples of ECE being applied to shallow foundations. Such an application would be feasible, albeit difficult. Galvanic protection and cathodic protection can also be used to reduce corrosion rates. ECE prolongs the service life of reinforced concrete by removing chloride ions from the concrete. As shown in Figure 10, ECE is performed by applying an anode mesh and electrolyte foam on the surface of the concrete. Upon applying a positive charge to the anode, chloride ions near the rein- forcing steel will migrate toward the anode, flowing through the electrolyte. Reductions of chloride ions of between 20% and 50% are common (American Concrete Institute 2001), reducing overall FIGURE 9 Pile repair decision matrix by Wan et al. (2013) for Wisconsin DOT.

21 corrosion potential. Chloride ions remaining in the concrete are further from the reinforcing steel, providing further service life gains. Load caPacity FHWA’s Foundation Characterization program identified load capacity issues as another of the four major challenges for foundation reuse. This section reviews how load capacity is evaluated for existing foundations, particularly compared with evaluations for new foundations, and methods for improving the load capacity of existing foundations for reuse applications. evaluating Load capacity of existing foundations Determining the load capacity of an existing foundation is considerably different from predicting the capacity of a new foundation for two primary reasons: 1. Difficulties investigating existing foundations add significant uncertainty to capacity inputs that are routinely taken for granted in predictions for new foundations. Such inputs include foundation geometry, depth, and material properties. 2. Historical information regarding applied loads and observed performance provide a lower bound on available capacity. Available methods for determining the load capacity of an existing foundation are listed and described here. Despite the differences described earlier, the available methods for existing founda- tions are generally similar to methods for new foundations, with the exception of the first method: • Identify capacity value from original bridge project documentation: plan drawings for bridges typically list design bearing pressures for shallow foundations and design loads for deep foundation elements. Knowledge of the design factor of safety (or resistance factor) is necessary to determine the ultimate foundation resistance from the original design. The design factor of safety might be ascertainable from the date and historical agency guidelines; however, discovery of the original design calculations is preferable. on discovery of the calculations it is also necessary to check the original calculations, which is recommended in the RuFUS manual (Butcher et al. 2006). For driven pile foundations, an additional estimate of load capacity can be made using a pile dynamic formula(s) if pile driving logs are available. • Presumptive capacity: Available load capacity is assumed based on information regarding the soil type and type of foundation. (a) (b) FIGURE 10 Application of ECE to substructure for reuse project in Richmond, Virginia: (a) anode mesh; (b) application of electrolyte (Source: Hardee 2014).

22 • Static (or rational) analysis methods: The methods most often used to predict load capacity of new foundations (e.g., bearing capacity theory and settlement analysis for shallow founda- tions, methods for predicting unit side and unit tip resistance values for deep foundations) can be applied to existing foundations as well. • High-strain dynamic methods (ASTM D495 2012) are a viable option for deep foundations. The methods are an appealing option if a pile driving hammer will be on site for the addition of driven piles and there is access to the heads of the existing piles with adequate headroom for the pile driving equipment. • Numerical methods: Finite element models of existing foundations can be useful, especially for predicting interactions with proposed new superstructures. • Load testing: It is possible to perform load tests of existing foundations, although physical constraints associated with the existing structure or other nearby infrastructure likely compli- cate testing. The RuFUS manual notes that it is sometimes possible to use the existing super- structure for reaction forces. Bell et al. (2013) presented a unique method for evaluating the capacity of existing foundations: the group installed fiber optic strain and temperature sensors along existing drilled shafts supporting an eight-story building in London and then monitored the foundation loads during the demolition of the building. The “unload test” resulted in profiles of strain, heave, and load in the shaft after six stories had been removed (Figure 11). The method is similar to measuring load-deflection response during the unload portion of a conventional static load test. The RuFUS manual includes a rationale for increasing the load capacity of an existing foundation from its original design value. The methodology is based on “reserves,” essentially capacity issues that are typically neglected out of conservatism during the original design. The RuFUS manual cautions that reserves cannot be assumed, but rather must be “carefully researched, assessed, and validated.” The manual organizes reserves into three categories: • overdesign reserves: – Applied load: Any documented applied loading with satisfactory performance can be used to establish an allowable loading in the future. – Strain reserve: Historical designs were often performed on an allowable stress basis. Using an allowable strain approach often increases the design resistance, particularly for founda- tion elements that exhibit relatively stiff load-displacement behavior. – over-capacity: often ground conditions or constructability considerations lead to the load capacity of an existing foundation being significantly greater than was necessary for the original design loading. – Historical constraints: previous standards for design may have been overly conservative. FIGURE 11 Unload monitoring profiles of (a) strain, (b) vertical heave, and (c) load during demolition of a 12-story building (Source: Bell et al. 2013).

23 • Reserve capacity: For deep foundation elements tied into walls, pile caps, or rafts the effects of the adjoining structures were often neglected or considered conservatively in the original design. • Time-related capacity: The capacity of many foundations, particularly driven piles in cohesive soils, increases with time. improving Load capacity of foundation Systems for reuse If the load capacity of the existing foundation system is insufficient for the proposed reuse loading, there are several methods of modifying the existing foundation system or the ground surrounding it that can be used to improve the load capacity. Examples of improving load capacity are included with the project examples at the end of this chapter. • Increasing footing size: FHWA’s Seismic Retrofitting Manual for Highway Structures: Part I— Bridges (Buckle et al. 2006) provides guidance for strengthening shallow foundations by increasing the footing size. The manual notes that increases in footing size are often accompa- nied by improvements to increase the structural capacity of the footing. The manual includes example design details for increasing footing size as well as example design analyses, although the analyses are focused on seismic issues. • Addition of deep foundation elements: Micropiles, driven piles, drilled shafts, ground anchors, or other deep foundation elements can be installed for additional load capacity. FHWA’s Seis- mic Retrofitting Manual notes that the addition of deep foundation elements is typically accom- panied by an increase in footing (pile cap) size as discussed previously. The RuFUS manual emphasizes that geotechnical and structural engineers must coordinate closely regarding the location of new elements and details for tying into the existing foundation system. • Ground improvement: The RuFUS manual identifies two mechanisms for increased load capacity by ground improvement. The first, “global ground modification,” involves ground improvement methods that improve the strength and stiffness of the overall soil or rock mass in which the exist- ing foundation is installed. Global ground modification techniques are also addressed in FHWA’s Seismic Retrofitting Manual, which primarily addresses ground improvement for the purpose of preventing liquefaction. The second mechanism described in the RuFUS manual, “pile improve- ment techniques,” involves ground improvement methods that are more narrowly targeted toward increasing the side and/or tip resistance of the individual deep foundation elements. The RuFUS manual presents several pile improvement techniques, but recommends shaft grouting to improve side resistance and jet grouting near the base of a deep foundation to improve tip resistance. • Replacement of backfill with lightweight fill: Technically, this option reduces the dead load on the existing foundation rather than increasing its load capacity; however, the effect on stability is similarly beneficial. This list is not exhaustive. There are various project-specific modifications that can be used to improve capacity or reduce required resistance. For example, Li et al. (2010) describe a bridge widening in Toronto for which the deck area was increased 40% without additional piles or widened footings. The existing bridge consisted of two twin structures. For the widening, the pile groups supporting the structures were tied together to create one rigid, combined pile group with significantly greater moment capacity than the sum of the two groups. The combined group approach also takes advantage of the reserve capacities of some piles in the individual groups, consistent with the recommendations in the RuFUS Manual described in the previous section. deciSioN MethodS Several methods originating in Europe have been developed to aid engineers and owners and agen- cies in decisions to reuse, retrofit, or replace existing foundations. A method included in the RuFUS manual (Butcher et al. 2006) is shown in Figure 12. The method is a qualitative assessment based on a series of yes-or-no questions contained within the flowchart. Examination of the primary questions on the left side of the flowchart shows close coordination with the major topics in the RuFUS manual, which overlap with the topics in this chapter. For instance, the fourth question from the top regard- ing the desk study and preliminary investigations refers to concepts presented in the Investigation and Condition Assessment section. If the project team answers “yes” to each of the questions on the flowchart, reuse is recommended.

24 FIGURE 12 RuFUS decision method for reuse of deep foundations (Source: Butcher et al. 2006).

25 United Kingdom-based engineering consulting firm Arup developed the Sustainable project Appraisal Routine (SpeAR®) diagram for reuse evaluation (Strauss et al. 2007). Example diagrams are shown in Figure 13 (Strauss et al. 2007). The diagram considers eight “drivers” for foundation reuse, with each driver represented by a segment on the diagram. A score from 1 to 6 is assigned to each driver, with lower scores indicating a greater value for reuse over replacement. Thus, the appeal of reuse increases as a site’s diagram more closely resembles a bullseye. Strauss et al. pro- vide descriptions for each driver, but no definitive guidance for how to assign a score to each. An example evaluation for a hypothetical project in Dublin, Ireland, is shown in Figure 13. Laefer (2011) explained the historical constraints and high land value of downtown Dublin that provided the motivation for reuse; however, other drivers were relatively weak. For example, the “nearly sub- urban” nature of much of Dublin corresponds to more than 10% of land being undeveloped, earning a score of 6 for “site location on previously developed land.” Laefer (2011) modified the SpeAR diagram method to include clear guidance for each driver as well as scores out of 6, not 7. For instance, the geologic conditions and constraints driver is assigned from 1 to 6 by matching site conditions with one of six descriptions provided in a table by Laefer. Laefer developed similar score tables for each of the other seven drivers. Laefer and Farrell (2015) proposed a hybrid reuse evaluation that includes the modified SpeAR diagram method as well as a modified version of the RuFUS flowchart. The RuFUS modification added several questions to the flowchart to incorporate related concepts included in another evalu- ation method from the CIRIA report (Chapman et al. 2007). Laefer and Farrell explained that the motivation for the hybrid approach was that the modified SpeAR method considers mostly socio- economic factors, whereas the RuFUS method considers mostly technical factors. Laefer and Farrell applied the hybrid approach to two reuse case examples and concluded that the hybrid approach successfully included consideration of all significant project constraints, although further testing and likely modification are necessary to address “scalability and robustness.” NotaBLe exaMPLeS of fouNdatioN reuSe There are many interesting and useful examples of foundation reuse. Three are presented in this section: (1) a project involving a series of accelerated bridge replacements, (2) an emergency repair project, and (3) a bridge widening project with modification of an existing foundation. (a) (b) FIGURE 13 (a) SPeAR diagram for foundation reuse evaluation and (b) SPeAR diagram for a project in Dublin, Ireland, in 2010 (Source: Laefer 2011).

26 fast 14 (Massachusetts dot) In the summer of 2011, the Massachusetts DoT (MassDoT) replaced superstructures for 14 bridges on I-93 in Medford. The bridges were founded on shallow foundations and steel H-piles. As shown in Figure 14, the accelerated bridge construction (ABC) project utilized technology involving pre- fabricated bridge elements and systems (pBES). ABC is a FHWA initiative designed to reduce the mobility impacts of construction projects using technology innovations such as pBES. For the Fast 14 project, all work was completed over weekends, with no impact on weekday traffic. Minimizing mobility impacts was valued because the interstate carries 200,000 vehicles per day in each direction. A similar series of bridge replacements was completed by Virginia DoT along I-95 in Richmond. Eleven superstructures were replaced with pBES and restoration work was completed on all sub- structures, including electrochemical chloride extraction for some piers and abutments (Collin and Jalinoos 2014). emergency repair of North carolina dot Bridge 91 Bridge No. 91 was built in 1980 and carries SR-1001 over the Yadkin River in central North Carolina. In May 2005, one of the bridge piers was observed to have tilted (Figure 15a), resulting in a span sepa- ration of nearly 3 inches. North Carolina DoT (NCDoT) closed the bridge and initiated an emergency investigation and repair. The investigation revealed through borings that rather than being founded on rock as designed the spread footing was bearing in alluvial materials. The emergency repair consisted of installing micropiles through the footings for the problem bent (Figure 15b), as well as for footings supporting two adjacent piers. The bridge reopened 3 months after closure. The emergency repair is an example of the difficulty of defining foundation reuse, which was discussed in chapter one of this report. By some interpretations, NCDoT’s emergency repair may not qualify as foundation reuse using the definition used in this Synthesis because the foundation design load was unchanged; how- ever, the project involved a critical decision regarding whether to replace the foundation altogether or to “reuse” it with improvements. In addition, the techniques used to investigate the existing foundation and to improve its capacity are techniques frequently applied for foundation reuse projects. henley Street Bridge widening (tennessee department of transportation) In 2011, Tennessee DoT began a project to widen, perform structural repair work, and seismic retrofit the historic Henley Street Bridge near Chattanooga. Some of the proposed work is shown in Figure 16a. The original bridge foundation consisted of groups of driven piles, which were deemed acceptable for reuse based on underwater inspection. To increase the capacity of the foundation system, drilled shafts were added between pile groups as shown in Figure 16b. The shafts were designed to carry most of the load, with the existing foundations supporting 20% of the total. FIGURE 14 Placement of PBES superstructure for one of the Fast 14 bridges (Source: Connors 2014).

27 (a) (b) FIGURE 15 Schematic representation of (a) foundation settlement resulting in rotation of the bridge pier and (b) micropile repair to stabilize the bent (Source: Mulla 2014). SuMMary of SigNificaNt fiNdiNgS • Structurally deficient bridges represent 10% of NBI bridges; functionally obsolete bridges represent 14%. Agency actions to manage bridge inventories generally consist of preventive maintenance, rehabilitation, or replacement. • In Europe, historical examples of foundation reuse date back centuries, where the idea has recently gained favor because of the difficulties of foundation construction in congested urban environ- ments. A European guidebook for foundation reuse, the RuFUS manual (Butcher et al. 2006), is available. U.S. experience may be more limited; however, a preliminary survey by NCDoT noted that agency experience with reuse is common. • Four major challenges for foundation reuse were identified by FHWA’s Foundation Character- ization program: condition assessment, load capacity, remaining service life, and design codes.

28 An additional challenge is an uncertain standard of care for consulting engineers involved in foundation reuse (Brown 2014). • Investigation of existing foundations intends to identify primary details (type of founda- tion, material type, dimensions, location, and depth) as well as structural integrity (condition assessment). • Many forms of historical records may provide value to the investigation of existing founda- tions. plan drawing sets are the most common and can help identify primary details. Historical inspection reports provide condition assessment data. The historical records can be used to scope the field investigations. • Field investigation methods reviewed included excavation and probing, concrete core drilling and laboratory testing, and geophysical methods. The advantages and limitations of each type of investigation were reviewed. The review indicated that it is possible to gather existing foun- dation information that is useful for design and evaluation of foundation reuse. The review also indicated that many of the methods, particularly the geophysical methods, are useful only for certain types of structures, foundations, and geologic conditions. • Deterioration mechanisms for steel, concrete, and timber affect the service life of foundations, as does bridge scour. Several cases of severe corrosion of steel piling have been documented in the last 30 years. The cases and guidance indicated that steel piling is typically only sus- ceptible to severe corrosion in aggressive environments associated with marine sites and sites with contaminated fill, generally from incineration byproducts, above the groundwater table. AASHTo Standard R 27-01 (2015) includes a procedure for assessing corrosion risks for steel piling in nonmarine environments. • Despite growing interest in service life, there is a lack of specific information available regard- ing how to predict foundation service life; current guidance simply recommends use of durable materials. AASHTo Standard R 27-01 (2015) does include a method for predicting the service life of steel piling if observations of pile condition and rate of corrosion are available. • Determining the load capacity of existing foundations is complicated by difficulties associated with the investigation of existing foundations; however, historical performance data provide information generally not available for the design of new foundations. Methods for determining existing foundation capacity are generally similar to those for new foundations. The RuFUS manual provides a basis for increasing capacity values from the original design values based on consideration of the conservatism often included in new design estimates. Monitoring existing foundations during demolition can provide a capacity estimate as well. • Methods for improving the capacity of an existing foundation generally consist of increasing the size of shallow foundations, the addition of deep foundation elements, ground improve- ment, or replacing backfill with lightweight fill. • The RuFUS foundation reuse decision flowchart evaluates technical aspects of reuse, whereas the modified SpeAR diagram evaluates socio-economic factors. (a) (b) FIGURE 16 (a) Henley Street Bridge with proposed widening and repairs; (b) proposed drilled shafts to be placed between existing pile groups (Source: Haddad 2014).