National Academies Press: OpenBook

Service Life of Culverts (2015)

Chapter: CHAPTER THREE Degradation Mechanisms

« Previous: CHAPTER TWO Synthesis of the State of the Practice
Page 10
Suggested Citation:"CHAPTER THREE Degradation Mechanisms." National Academies of Sciences, Engineering, and Medicine. 2015. Service Life of Culverts. Washington, DC: The National Academies Press. doi: 10.17226/22140.
×
Page 10
Page 11
Suggested Citation:"CHAPTER THREE Degradation Mechanisms." National Academies of Sciences, Engineering, and Medicine. 2015. Service Life of Culverts. Washington, DC: The National Academies Press. doi: 10.17226/22140.
×
Page 11
Page 12
Suggested Citation:"CHAPTER THREE Degradation Mechanisms." National Academies of Sciences, Engineering, and Medicine. 2015. Service Life of Culverts. Washington, DC: The National Academies Press. doi: 10.17226/22140.
×
Page 12
Page 13
Suggested Citation:"CHAPTER THREE Degradation Mechanisms." National Academies of Sciences, Engineering, and Medicine. 2015. Service Life of Culverts. Washington, DC: The National Academies Press. doi: 10.17226/22140.
×
Page 13
Page 14
Suggested Citation:"CHAPTER THREE Degradation Mechanisms." National Academies of Sciences, Engineering, and Medicine. 2015. Service Life of Culverts. Washington, DC: The National Academies Press. doi: 10.17226/22140.
×
Page 14
Page 15
Suggested Citation:"CHAPTER THREE Degradation Mechanisms." National Academies of Sciences, Engineering, and Medicine. 2015. Service Life of Culverts. Washington, DC: The National Academies Press. doi: 10.17226/22140.
×
Page 15
Page 16
Suggested Citation:"CHAPTER THREE Degradation Mechanisms." National Academies of Sciences, Engineering, and Medicine. 2015. Service Life of Culverts. Washington, DC: The National Academies Press. doi: 10.17226/22140.
×
Page 16
Page 17
Suggested Citation:"CHAPTER THREE Degradation Mechanisms." National Academies of Sciences, Engineering, and Medicine. 2015. Service Life of Culverts. Washington, DC: The National Academies Press. doi: 10.17226/22140.
×
Page 17
Page 18
Suggested Citation:"CHAPTER THREE Degradation Mechanisms." National Academies of Sciences, Engineering, and Medicine. 2015. Service Life of Culverts. Washington, DC: The National Academies Press. doi: 10.17226/22140.
×
Page 18
Page 19
Suggested Citation:"CHAPTER THREE Degradation Mechanisms." National Academies of Sciences, Engineering, and Medicine. 2015. Service Life of Culverts. Washington, DC: The National Academies Press. doi: 10.17226/22140.
×
Page 19
Page 20
Suggested Citation:"CHAPTER THREE Degradation Mechanisms." National Academies of Sciences, Engineering, and Medicine. 2015. Service Life of Culverts. Washington, DC: The National Academies Press. doi: 10.17226/22140.
×
Page 20

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.

8 CHAPTER THREE DEGRADATION MECHANISMS This chapter summarizes the degradation mechanisms that cause deterioration in the serviceability of culvert pipes over time. The environmental, structural, and hydraulic load- ing conditions that lead to degradation are also addressed. Corrosion and abrasion are the two primary degradation mechanisms for properly specified and installed culvert pipe systems. These two aspects will be addressed in separate subsections; however, it is important to note that corrosion and abrasion are processes that work in tandem and may cause a combined effect greater (more detrimental) than simply the combined sum of each process applied separately. Discussion on this combined effect is presented in the sec- tion on the combined effect of corrosion and abrasion. The other forms of nonpressure pipe degradation can be described as weathering effects. These include damage as a result of freeze-thaw cycles, slow crack growth, and expo- sure to ultraviolet (UV) radiation. Section 12.6.9 of the LRFD [Load and Resistance Fac- tor Design] Bridge Design Specifications (AASHTO 2013) requires that the degradation of structural capacity result- ing from corrosion and abrasion be considered in design, but does not provide specific methods for doing so. The specification further allows that if the design of a metal or thermoplastic culvert is controlled by flexibility factors (i.e., construction loads versus service loads) during installation, then the requirements for corrosion and abrasion protection may be reduced or eliminated, provided that it is demon- strated that the degraded culvert will provide adequate resis- tance to loads throughout the service life of the structure. CORROSION Corrosion is the loss of section or coating by chemical or electrochemical processes (AASHTO 2010). Corrosion most commonly impacts metal culverts or the metal reinforce- ment in concrete pipe. Figure 7 schematically depicts the mechanisms and life cycle of metal corrosion. All corrosion processes involve the flow of current from one location to another (a corrosion cell) (AISI 1999). As such, corrosion requires the presence of water or some other liquid to act as an electrolyte, with pipe materials acting as an anode, cathode, or conductor. As electrons move from the anode to the cathode, metal ions are released into solution. This causes characteristic pitting at the anode. In culvert pipe applications, the culvert itself will typically serve as both the anode and the cathode. A summary table and sche- matic of common corrosion mechanisms after ASM (for- merly American Society of Metals) is provided in Figure 8. FIGURE 7 Life-cycle schematic of metal corrosion (after M. Paredes, FDOT, personal communication, May 5, 2014). Corrosion can affect either the inside (water side) or out- side (soil side) of a pipe or both. The potential for corrosion to occur, and the rate at which it will progress, is dependent on a variety of factors, including: • pH • Resistivity • Chlorides • Sulfates • Other conditions (soil moisture content, dissolved gases, bacterial activity, etc.). Depending on the particular nature of the corrosive envi- ronment, the following mitigation measures may be required: • Increased wall thickness (metal culverts) • Additional cover over reinforcing steel (concrete culverts) • Coatings or protective pavings applied to the culvert (all culvert material types) • Electrical grounding or cathodic protection, or both • Placement of the culvert in a nonaggressive (e.g., gran- ular) backfill.

9 pH pH is a measure of a solution’s acidity or alkalinity. It is a mea- sure of the concentration of hydronium ions in solution, and ranges from 0 to 14. Acidic solutions have pH values less than 7, and alkaline (or basic) solutions have pH values greater than 7. A solution with a pH of 7 is considered neutral. pH values in natural waters generally fall within the range from 4 to 10. A pH value less than 5.5 is consid- ered to be strongly acidic, while values of 8.5 or greater are considered to be strongly alkaline. pH values that are either highly acidic or highly alkaline are indicative of an increased potential for corrosion. Generally, pH levels between 5.5 and 8.5 are not considered to be severely det- rimental to culvert life. The lowest pH levels in natural soils are typically seen in areas that have received historically high rainfall where the runoff and percolation have leached soluble salts from the soil, resulting in the soil becoming acidic. Other likely sources of potentially acidic runoff are from naturally occurring acid-generating geologic formations, mine sites, and other industrial wastes. Milder acids can be found in runoff from marshy areas, which contain humeric acid, and mountain runoff that may contain carbonic acid. Arid areas are more likely to be alkaline owing to soluble salts con- tained in groundwater being drawn to the surface through capillary action and then concentrating in the soil after water evaporation occurs through the normal daily and seasonal drying cycles. Resistivity Resistivity of soil is a measure of the soil’s ability to conduct electrical current. It is affected primarily by the nature and concentration of dissolved salts; the temperature, moisture content, and compactness of the soil; and the presence of inert materials such as stones and gravel. The greater the resistivity of the soil and/or the lower the soil moisture con- tent, the less capable the soil is of conducting electricity and the lower the corrosive potential. FIGURE 8 Table and schematics of common corrosion mechanisms (after ASM International 2003).

10 Resistivity values in excess of 2,000 to 5,000 ohm-cm (depending on the reference guideline) are generally consid- ered to present limited corrosion potential (Table 1). Resis- tivity values below the range of 1,000 to 3,000 ohm-cm will usually require some level of pipe protection, depending on the corresponding pH level and pipe material susceptibil- ity to corrosion. In general, the lower the pH, the higher the resistivity at which mitigation measures may be required. TABLE 1 TYPICAL RESISTIVITY RANGES FOR SOIL AND WATER Classification Resistivity (ohm-cm) Water Surface water Brackish water Seawater R > 5,000 R = 2,000 R = 25 Soil Rock Sand Gravel Loam Clay R > 50,000 50,000 > R > 30,000 30,000 > R > 10,000 10,000 > R > 2,000 2,000 > R > 750 Sources: After NCHRP Synthesis Report 254 and AISI (1999). As a comparative measure, resistivity of seawater is in the range of 25 ohm-cm, clay soils range from approximately 750 to 2,000 ohm-cm, and loams range from 3,000 to 10,000 ohm-cm. Soils that are of a more granular nature typically exhibit even higher resistivity values and as such present lower risk to resistivity induced corrosion (Tables 2–4). TABLE 2 TYPICAL SOIL CORROSION POTENTIAL RESISTIVITY VALUES Soil Corrosion Potential Resistivity (ohm-cm) Negligible Very Low Low Moderate Severe R > 10,000 10,000 > R > 6,000 6,000 > R > 4,500 4,500 > R > 2,000 2,000 < R Sources: After NCHRP Synthesis Report 254 [Gabriel and Moran (1998)]. TABLE 3 TYPICAL SOIL CORROSION POTENTIAL RESISTIVITY VALUES Soil Corrosion Potential Resistivity (ohm-cm) Normal Mildly Corrosive Corrosive R > 2,000 2,000 > R > 1,500 1,500 > R Sources: After AISI (1999). Chlorides Dissolved salts containing chloride ions can be present in the soil or water surrounding a culvert. Chlorides will also be of concern at coastal locations, near brackish water sources, and at locations that use winter deicing salts. TABLE 4 TYPICAL CORROSION POTENTIAL OF VARIOUS SOIL CONDITIONS Soil Type Description of Soil Aeration or Drainage Water Table 1—Lightly Corrosive • Sands or sandy loams • Light-textured silt loams • Porous loams or clay loams thoroughly oxidized to great depths Good Very low 2—Moderately Corrosive • Sandy loams • Silt loams • Clay loams Fair Low 3—Badly Corrosive • Clay loams • Clays Poor 2 to 3 ft below surface 4—Unusually Corrosive • Muck • Peat • Tidal marsh • Clays and organic soils Very poor At surface or extreme imperme- ability Source: After Hurd (1984). In most instances, corrosive potential increases as the negative chloride ion decreases the resistivity of the soil or water and destroys or degrades protective films on anodic areas. Chlorides, as with most of the more common corro- sive elements, primarily attack unprotected metal culverts and the reinforcing steel in concrete culverts if the concrete cover is inadequate, cracked, or highly permeable. Sulfates Sulfates can occur naturally or may result from human activity, for example, agricultural runoff, mine wastes, ille- gal dumping effluents, and spills. Sulfates, in the form of hydrogen sulfide, can also be created from biological activ- ity, which is more common in wastewater, sanitary sewers, and some industrial piping applications, and can combine with oxygen and water to form sulfuric acid. Although high concentrations of sulfates can lower pH, and be of concern to metal culverts, sulfates are typically more damaging to concrete culverts. Typically, sulfates (in various forms) combine with the lime in cement to form cal- cium sulfate (gypsum), which creates structural weakness in concrete culverts and promotes degradation. Concrete pipe can normally withstand sulfate concentra- tions up to 1,000 parts per million without special consider- ations. For higher concentrations of sulfates, higher-strength concrete, concrete with lower amounts of calcium aluminate, or special coatings may be necessary.

11 Microbially Induced Corrosion Corrosion promoted or caused by microorganisms is known by a number of different terms, including; microbially induced corrosion, microbial corrosion, bacterial corrosion, biocorrosion, and microbiologically influenced corrosion. The term microbially induced corrosion (MIC) will be used throughout this report. In this report, the term MIC will also refer to both the direct and indirect effects that microorgan- isms have on corrosion. MIC is the deterioration of metals resulting from the met- abolic activity of microorganisms, and has been identified as one of the major causes of corrosion failures of buried metal structures. MIC primarily affects metal culverts but can also affect the reinforcing steel in reinforced concrete culverts. Many industries are affected by MIC, primarily those in marine and coastal environments. As part of the environ- mental characterization of a highway drainage project site, factors relevant for MIC are now being investigated (Sagüés et al. for FDOT 2009). It has been reported (Peng and Park 1994) that almost half of Wisconsin’s steel culvert corrosion was related to MIC. MIC can occur in many metals, including carbon steel, stainless steel, aluminum alloys, and copper alloys. MIC can occur in pH ranges from approximately 4 to 9, and in tem- peratures ranging from approximately 10°C to 50°C. MIC presents as corroded metal surfaces covered in slime, black iron sulfide deposits, algal growth, and as a rotten-egg odor. Microorganisms’ actions can either inhibit or promote corrosion by changing the corrosion reactions that occur at the metal’s surface. Microorganisms also affect the for- mation of biofilms, which in turn can also inhibit or pro- mote corrosion by changing the pH, acting as a catalyst for corrosion reactions, acting as a barrier to gas diffusion, and harboring other microorganisms that may influence MIC reactions. Many microorganisms are thought to influence MIC, including iron-oxidizing, sulfur-oxidizing, iron-reducing, and sulfur-reducing microorganisms. Sulfur-reducing bac- teria are widely believed to be largely responsible for MIC in anaerobic conditions. MIC reactions are generally localized and occur at cracks, crevices, and areas where the metal has been welded. Other factors that influence the rate of MIC are the availability of oxygen and organic carbon, with an increase in availability of these two components causing an increased rate of MIC. Based on a field study by Sagüés et al. (2009) for the Florida DOT (FDOT), the following general observations regarding MIC of metals used for highway drainage pipes can be made: • Carbon steel, galvanized steel, and aluminized steel are all susceptible to MIC. • The potential for MIC is reduced where pipe flow is rapid and the pipe is placed above the water table in free-draining soils or engineered backfill. • Consideration should be given to determining the organic carbon content of the soil and water to assess the potential for MIC. Other Corrosion Considerations Industrial Effluent Industrial effluents can contain compounds that are extremely destructive to pipe materials. Waste streams from most industries are sufficiently regulated to be of limited concern to the highway engineer. However, tailings from historic (i.e., less regulated) mining operations (or natural runoff from minable geologies) can be a source of highly acidic runoff, as can livestock operations or illegal connec- tions from residential or small commercial lots. Potentially corrosive runoff can also be of concern at locations known for a high probability of accidental spills (e.g., runaway truck escape ramps). An assessment of the presence and concentrations of cor- rosive constituents in the streamflow needs to be conducted whenever industrial effluents are suspected in the runoff. If the source can be identified, corrective action can usually be taken or culvert protective measures can be implemented. Stray Electrical Current Corrosion can be induced by electric current in proximity to the pipe. Although corrosion most often affects metal pipes, the steel in reinforced concrete pipes may also suffer an increased rate of corrosion. Typical sources of stray current are electrified rail lines, high-tension electric transmission lines, and cathodically protected structures (gas transmis- sion mains). Protective coatings are usually applied to the pipe to negate the effects of stray electric currents. ABRASION Abrasion is the progressive loss of section or coating of a cul- vert by the continuous, rapid movement of turbulent water containing a bedload of particulate matter (sands, gravel, transported debris, etc.). Abrasion will almost always mani- fest itself first in the invert of the culvert. As with corrosion, several factors contribute to abrasive potential, including culvert material, frequency and velocity of flow in the cul- vert, and bedload composition. AASHTO (2007) Chapter 14 advises against the use of metal pipe in abrasive environments unless the invert is

12 paved. Ault and Ellor (2000) and NCHRP 10-86 (2015) rec- ommend incorporating the existing Federal Lands Highway Design Guidance abrasion rating system (Levels 1 through 4) into culvert condition assessment and durability predic- tion practices at a minimum. Bedload Bedload is the portion of the total transported sediment that is carried by intermittent contact with the streambed (or culvert invert) by rolling, sliding, and bouncing. Contact between bedload and the culvert pipe is the leading cause of culvert abrasion. Critical factors in evaluating the abrasive potential of bedload are the size, shape, and hardness of the bedload material, and the velocity and frequency of flow in the culvert. Flow velocities depend on the drainage barrel roughness, the cross-sectional geometry, slope, and the depth of flowing water. Abrasion will increase by a factor of approximately four when the flow velocity is doubled. Theoretically, dou- bling the velocity of a stream increases its ability to transport solid fragments of a given size by as much as a factor of 32. Abrasion is thus highly sensitive to the flow velocity. The AASHTO Highway Drainage Guidelines (2007) define bedload by the 2- to 5-year return frequency flow velocity. Generally, flow velocities less than 5 ft/s are not considered to be abrasive, even if bedload material is pres- ent. Velocities that exceed 15 ft/s and carry a bedload are considered to be very abrasive. Tests performed on concrete pipe have generally shown excellent wear characteristics with respect to abrasion resis- tance. Although high-velocity flow will induce abrasion regardless of the size of bedload particles, tests performed on concrete pipe have shown that cobble and larger sizes will induce higher wear rates than sands and gravels. Larger rocks strike with enough force to break away small particles of the concrete pipe wall. The use of high-quality aggregate (i.e., aggregate that is harder than the anticipated bedload hardness) in the concrete mix can greatly enhance the con- crete’s resistance to abrasion. Manufacturing methods that lead to a denser concrete mix, such as roller-compacted or spun concrete or higher- compressive-strength concrete, can also exhibit increased resistance to abrasion. Where velocities are known to be high, and a bedload is present, many agencies recommend additional concrete cover over the reinforcing steel. Debris Debris carried by storm waters can also be a destructive element in culverts. However, this destructive potential is primarily related to clogging of the culvert by the attendant effects of overtopping and erosion or to a single impact from a large piece of debris that causes immediate damage to the culvert. Large volumes of debris can, however, add to the effects of bedload abrasion. The potential for debris to add to abrasion will depend primarily on the relative hardness of the debris and the culvert material. The most common types of debris that lead to major damage are boulders, trees and shrubs, and ice, although during major storm events, anything movable by storm waters can be transported to culvert locations. Types of areas that have proven troublesome are drainages with unstable hillsides, heavily forested areas subject to fire, streams that support beaver activities, and cold-weather sites where ice accumulation can block or otherwise dam- age drainage structures. Whenever debris is likely to pose a problem, appro- priate debris-control structures should be considered for installation. FHWA Definitions of Abrasion Levels The following abrasion levels are intended as guidance to help the engineer consider the impacts of bedload wear on the invert of pipe materials. Sampling of the stream- bed materials is not required, but visual examination and documentation of the size of the materials in the streambed and the average slope of the channel will give the designer guidance on the expected level of abrasion. Where existing culverts are in place in the same drainage, the conditions of inverts could also be used as guidance. The expected stream velocity should be based on a typical bank-full design dis- charge generated by a 2- to 5-year return frequency flood and not a 10- or 50-year design flood. • Level 1. Nonabrasive conditions exist in areas of no bedload and very low velocities. This is the condition assumed for the soil side of drainage pipes. • Level 2. Low abrasive conditions exist in areas of minor bedloads of sand and velocities of 5 ft/s or less. • Level 3. Moderate abrasive conditions exist in areas of moderate bedloads of sand and gravel and velocities between 5 ft/s and 15 ft/s. • Level 4. Severe abrasive conditions exist in areas of heavy bedloads of sand, gravel, and rock and velocities exceeding 15 ft/s. Caltrans Definitions of Abrasion Levels The California DOT (Caltrans) Highway Design Manual, Chapter 850 (Caltrans 2011b), provides comprehensive guidance on abrasion levels coupled with material selection guidance and estimates of additional service life provided by various protective coatings. A series of tables provides specific guidance, as outlined in the following list:

13 • Table 855.2A: Definitions of abrasion levels and cor- responding material recommendations. • Table 855.2B: Bed-material size and estimate of non- scour velocities for different flow depths. • Table 855.2C: Estimated additional service life as a result of protective coatings. • Table 855.2D: Estimated wear (mils/year) for corru- gated metal pipe under different abrasion levels. • Table 855.2E: Relative assessment of abrasion-resistant materials. • Table 855.2F: Guide for minimum material thickness to achieve 50-year maintenance-free service life. DeCou and Davies (2007)—Caltrans Abrasion Study DeCou and Davies’ (2007) 5-year study on Shady Creek in Nevada County, California, is the most in-depth pipe abra- sion study found. Figure 9 shows the variety of pipe material and coating-type test specimens used in the California Abra- sion study. Taylor and Marr (2012) provide an excellent sum- mary of this site-specific abrasion study at a highly abrasive site, which is explained in the following text. FIGURE 9 Caltrans abrasion test panel installation showing various culvert materials and coatings (Caltrans 2013). This site, with average flow velocities of 12 to 18 ft/s and median grain sizes between 3 and 11 mm, is highly abra- sive. The service life estimates developed for this study are site specific because of these conditions. DeCou and Davies found that abrasive wear at the site is event driven and not linear with time. Several material comparisons and observa- tions were made: • All nonconcrete pipe materials studied have lower abrasive wear rates than concrete; however, concrete pipe walls are much thicker than the nonconcrete pipe materials studied. • Smooth pipes wear slower than rough-walled pipe. • PVC pipe wears slower than HDPE; however, the con- struction of smooth-walled, corrugated HDPE pro- vides a positive characteristic. After the inner wall is perforated, the outer wall remains intact. • Polyethylene coating for composite steel spiral rib pipe was the only steel coating studied that could provide the desired 50-year service life. COMBINED EFFECT OF CORROSION AND ABRASION The abrasive properties of bedload that is traveling at high velocities and is harder than the exposed pipe invert or coating will erode metal, concrete, and thermoplastic pipes. Erosion may begin with the formation of corrosion products of the pipe material. These corrosion products are often more brittle than the parent material from which they were formed and may then be removed by the bedload’s abrading action more easily than the parent material. The parent pipe material is then reex- posed and not protected against subsequent cycles of corrosion and abrasion. When corrosion and abrasion operate together in this manner, they can produce a larger detrimental effect than either would if applied in isolation. Abrasion accelerates corrosion by removing protective coatings, and corrosion can produce products less resistant to abrasion (Figure 10). FIGURE 10 Corrosion accelerated by abrasion causing void formation below (Caltrans 2013). Water flowing at a velocity high enough to create appre- ciable turbulence can also cause a localized effect known as impingement. Impingement is caused by suspended solid particles (as opposed to abrasion, which is caused by par- ticles transported along the streambed) or gas bubbles strik- ing the surface and can occur at pipe entrances, sharp bends, protrusions (such as rivets and lapped joints), and other abrupt changes in flow patterns. The protective layer of a metal or concrete can thus be locally compromised, facilitat- ing subsequent corrosion of an unprotected material. Steel culverts are the most susceptible to the dual action of abrasion and corrosion, particularly where thinner-walled pipes are used. Once a steel pipe’s thin protective coating—whether

14 it is zinc or another substance—is worn away, exposure to low- resistivity or low-pH environments can dramatically shorten a steel culvert’s life. Although aluminum culverts are occasion- ally specified to combat corrosion, plain aluminum is typically not recommended for abrasive environments since tests indi- cate that aluminum can abrade as much as three times faster than the rate of steel. Plastic culvert materials (both PVC and HDPE) exhibit good abrasion resistance. Since plastic is generally not subject to corrosion, it will not experience the dual action of corrosion and abrasion. Plastic pipes, like metal pipes, have relatively thin walls and thus the rate of wear must be carefully evalu- ated with the material thickness. The documented abrasive- resisting capabilities of plastic pipe is primarily based on tests using small aggregate sizes (gravels and sands) flowing at velocities ranging from 2 to 7 ft/s (AASHTO 2007). The effects of large bedload particles (cobbles and larger) or high- velocity flows are not well defined because of limited data. Additionally, as a result of their more recent emergence as a culvert product, plastic pipes have generally not had rehabili- tative strategies developed specifically for them. Some of the more common current strategies (e.g., invert paving) are not effective with plastic pipes because of their smooth surface and inability to achieve a satisfactory bond. An illustrative case study example of combined corro- sion and abrasion is provided by Caltrans (2013). It relates to the back-analysis of a structural steel plate pipe culvert with inlet velocity of 12 ft per sec and outlet velocity of 22 ft per sec. The original steel thickness was 0.140 in. (10 gage). From a back-analysis of the time to perforation, the rate of steel loss was estimated at about 4.6 mils per year. Using the site values for pH and resistivity, the contribution from cor- rosion alone was estimated at 2.7 mils per year, indicating that the contribution to metal loss from abrasion was about 1.9 mils per year. It was concluded that to provide a 50-year service life for that application with abrasion level 5, three- gage (0.250 in.) steel plate would have been needed. This culvert was replaced with reinforced concrete pipe (RCP) and within 5 years the steel was exposed, indicating that the concrete loss was approximately 200 mils per year. The ulti- mate solution was to provide 12 in. of concrete paving with a flat bottom to spread the flow concentration. OTHER DURABILITY FACTORS Some factors that can potentially impact culvert service life but are less common than the primary mechanisms of cor- rosion and abrasion are briefly summarized in this section. Freeze/Thaw Free-draining bedding requirements and extra care with pipe joint protection are needed when the top of the pipe (crown) is located above the frost-penetration depth; how- ever, the requirements are independent of pipe material type. Particular attention is needed for pipe replacements or exten- sions during road rehabilitation works to ensure that the frost treatment details are maintained uniformly between old and new construction. Typically, buried pipe is not exposed to freeze-thaw conditions when installed as a storm sewer below the frost-penetration depth. However, culverts are frequently installed within the frost zone and deeper instal- lations are exposed to frost action at the inlets and outlets. Failure to account for freeze-thaw impacts across all pipe materials may lead to differential settlement causing joint separation, longitudinal cracking of the pipe, localized over- stressing, and decreased hydraulic performance. It can also lead to differential performance of the road pavement above the pipe. The potential impacts of freeze-thaw are typically alleviated through the use of frost tapers (the incorporation of excess sloped excavations around culvert locations back- filled with free-draining granular backfill). High-Humidity Conditions High humidity (100% relative humidity) and high atmo- spheric temperatures (> 85°F/30°C) are not uncommon within gravity pipes, such as in swamp or marsh areas with partially submerged, stagnant, or low flow conditions. In such an environment, hydrogen sulphide released from stagnant, sewage-like conditions is absorbed by the film of moisture on that portion of the pipe lying above the water. In the presence of aerobic bacteria, the hydrogen sulphide is converted to sulphuric acid. This can lead to deteriora- tion of concrete and steel, although the pipe materials are not directly affected by humidity and temperature. Time-Dependent Mechanical Properties Thermoplastic materials (HDPE and PVC) are viscoelas- tic; that is, their mechanical properties are time-dependent and incur strain and creep deformation under a sustained load, or exhibit stress and load relaxation under a sustained deflection. HDPE and PVC pipes sustain deformations that are controlled by the surrounding soil, so stress relaxation in the pipe can be expected over its lifetime. Slow crack growth and oxidative and chemical failure have been identified as the primary long-term failure mechanisms for corrugated HDPE pipes. No methods based on service histories have yet been developed for serviceable life predictions for these materials; rather, material specifications are used to assign standard service life values based on historic performance or laboratory bench-scale evaluations. Figure 11 schematically illustrates the time-dependent oxidation mechanism for sta- bilized and unstabilized polyethylene (M. Paredes, FDOT, personal communication, May 5, 2014). Established practice (AASHTO 2013) is to account for the long-term material response by employing a long-term “effec-

15 tive” modulus of elasticity selected in accordance with the design life of the system (the longer the time period, the lower the modulus). Thermoplastics are relatively resistant to cor- rosion and abrasion in buried highway drainage applications; therefore, the effective modulus of elasticity may control the long-term stability (this material response over time can be considered one factor dictating the estimated material service life). This “modulus of relaxation” can be obtained experimen- tally by dividing a residual stress in the pipe wall by the strain at that location, and can be estimated from measurements made using constant deflection tests conducted on the pipe as part of quality assurance processes (Gabriel and Moran 1998). FIGURE 11 Schematic of time-dependent oxidation degradation mechanism for polyethylene (M. Paredes, FDOT, personal communication, May 5, 2014). Moore and Hu (1996) have examined the time-dependent behavior of various HDPE materials and provide viscoelastic parameters that can be used to estimate “relaxation moduli” at various time intervals. AASHTO (2014) provides short and 50-year values of modulus for both PVC and HDPE materials, and 100-year values have recently been proposed by McGrath and Hsuan (2004), updated in McGrath et al. (2009) (Figure 12). Slow Crack Growth Slow crack growth (SCG) occurs because thermoplastics, when subjected over a long time period to tensile stresses substantially lower than those necessary to cause a short- term rupture, can develop crazes and small cracks that grow slowly until eventually rupture occurs (NCHRP 14-19 2010). Crazes are very fine cracks that develop in the direction nor- mal to tensile stress; their surfaces are still bounded together by molecular fibrils, approximately 10 nm in diameter, which continue to support the load (i.e., crazes represent the redistri- bution of local stresses throughout the thermoplastic matrix). The formation of these crazes and cracks is not caused by any chemical degradation of polymer and is only the result of mechanical or thermal forces (McGrath et al. 2009). NCHRP 14-19 reports that, in general, the rate of SCG can be accelerated by different factors, for example, stress inten- sity, cycling of the stress (fatigue), elevated temperature, and exposure to certain environments (referred to as environmen- tal stress cracking). This type of brittle cracking in HDPE pipes generally results from a combination of high tensile stress (resulting from applied loads, residual stresses, or ther- mal effects) and low-quality resin with poor crack resistance. It may be associated with simple two-dimensional behavior of the pipe (where circumferential tensions develop on the outside surface of the pipe at the springlines, or on the inside surface at the crown or invert). More commonly, the tensile stresses result from three-dimensional behavior caused by complexities of the profile (e.g., Moore and Hu 1995). A series of studies were conducted for FDOT (FDOT 2008a–d), following which they developed a new specifica- FIGURE 12 Microscope images of HDPE crack surfaces (Hsuan and McGrath 1999): (left) flake morphology characteristic of impact-type fracture; (right) fibrous morphology characteristic of slow crack growth.

16 tion incorporating material performance and test require- ments to allow HDPE pipe materials to be given a default 100-year service life. The material requirements take into account the higher ambient temperatures present in Florida, but do not consider freeze-thaw issues and as such are some- what regionally specific in their applicability and acceptance across North American transportation agencies. Ultraviolet Radiation No reports indicate that UV radiation degrades concrete or steel. HDPE and PVC pipe may incur surface damage when exposed to long-term UV radiation, typically at the exposed ends of culverts. UV degradation may include color change, a slight increase in tensile strength and elastic modulus, and a decrease in impact strength. FDOT limits exposure of UV- susceptible pipe materials to 2 years (M. Paredes, FDOT, personal communication, May 5, 2014). With the use of carbon black (a UV stabilizer), HDPE pipe is protected against prolonged exposure to sunlight and the potential for UV degradation of mechanical properties. UV stabilizers are used in PVC pipe materials to pro- tect against UV degradation, although the longevity of these additives has not been proven. However, it is considered pru- dent to protect the exposed ends of installed PVC (and to a lesser extent HDPE) pipes that include UV stabilizers. Once buried, except for exposed ends, exposure of plastic pipe to sunlight generally does not occur. Exposure issues often can be overcome if nonsensitive (e.g., concrete or steel) end walls are used. Outdoor storage practices are to be managed by the manufacturers to ensure that the pipes are not subject to prolonged UV exposure prior to site delivery. Seismically Induced Degradation For small-span (less than 10 ft) gravity-pipe road applica- tions, seismic loads are generally not considered in design in many areas. However, for high-risk applications with the potential for upstream flooding or for permanent ground displacement, it is recommended that a seismic design be incorporated into the structural analysis. Seismic loads on installed pipes arise from inertia forces owing to earthquake shaking or from large permanent ground movement generally associated with strength and stiffness loss of loose or sensitive saturated foundation soils. Liquefac- tion and lateral spreading are the main causes of pipe failures, and these failure modes should be considered in design. Access/Construction Equipment Damage to pipes from overloading during construction is a common issue and can significantly reduce the service life of culverts. Most agencies set minimum fill heights above the crown of the pipe before power-operated tractors or rolling equipment can be used for compaction. The speci- fied minimum heights for heavy-equipment crossing may affect gravity-pipe selection and may require the placement of temporary fill protection for pipes during construction. Depending on final grade restrictions, fill-material costs, and construction traffic, these minimum fill heights may influence the initial installation costs and are to be consid- ered in the life-cycle cost analysis. Thermoplastic pipes are particularly vulnerable to dam- age during installation, hence the need for rigorous post- installation inspection. Examples of random-type damage to thermoplastic pipe as observed from video inspection are shown in Figures 13–15. FIGURE 13 Approximately 2-in.-diameter puncture in HDPE pipe. FIGURE 14 Split in inner pipe wall and local buckling. Impact Resistance (Brittleness) and Temperature Effects During construction, pipes are required to withstand forces that are normally expected during shipment, handling, and instal- lation. In addition, rockfill is often used above the pipe cover

17 material, and rock fragments that are used to form the side slopes and embankments will frequently roll onto exposed pipe ends or penetrate the pipe’s overlying cover or embedment soil. Temperature affects all pipe materials differently. For normal operating temperatures experienced during highway construc- tion projects in Ontario, the use of concrete, steel, high-density polyethylene, and polyvinyl chloride is considered acceptable. Special provisions should be made when pipe installations are required in rare extreme-temperature conditions (i.e., below -228°F/-308°C) for emergency situations. Table 5 can be used as a guideline for minimum installation temperatures. FIGURE 15 Wall penetration and puncture from metal spike. TABLE 5 RECOMMENDED MINIMUM PIPE INSTALLATION TEMPERATURES Pipe Material Minimum Installation Temperature1 (°F)/(°C) Concrete -22/-30 HDPE -22/-30 PVC2 0/-18 Steel -22/-30 Source: MTO (2007). Notes: 1Minimum operating temperature for workplace assumed to be -30°C. 2AASHTO (2000) reports PVC as becoming brittle at exposure to temperatures less than 37°F/3°C, and many agencies (e.g., Minnesota DOT) specify that brittle transition temperature as the minimum allowable during installation due to the risk of construction impact–based damages. The Ministry of Transportation of Ontario (MTO) design guidance (MTO 2007) provides the following additional commentary on temperature impacts. Concrete material compressive and tensile strengths are reported to increase with a reduction in operating temperatures. The effect of temperature on the impact strength of steel material is not considered an issue and the pipe itself can be designed to withstand handling and installation forces according to ASTM A796 (defined as the flexibility factor). HDPE mate- rial has an impact resistance that ranges from about 0.27 to 0.80 Nm/mm (Vasile and Seymour 1993); however, this can be reduced significantly by oxidation resulting from sun- light or by overheating during the manufacturing extrusion process. PVC material has an impact resistance much less than that of HDPE, of about 0.026 Nm/mm (Titow 1990), but can be increased to about 1.07 Nm/mm by blending with an impact modifier during the extrusion process. However, impact modifiers may reduce chemical resistance, increase susceptibility to oxidation, and increase permeability. Fire Damage Although the risk of damage to storm drainage systems is quite low, under certain circumstances, such as forest fires, damage to culverts can occur. In forest fires, all pipe mate- rial types can sustain damage from exposure to extremely high temperatures. While thermoplastic pipes would be the most vulnerable, the National Fire Protection Association (NFPA 2012) has given both polyethylene and polypropyl- ene a rating of 1 (Slow Burning) on a scale of 0 to 4, where higher ratings indicate a greater vulnerability. Existing “Pipe System” Conditions Where new pipe is to be installed and incorporated into an existing pipe system, an assessment of the existing pipe mate- rial type should be made prior to design. If the existing pipe material is considered to be performing satisfactorily, many agencies prefer to maintain the same pipe material to minimize the risk of construction and performance issues related to con- nections, joints, geometrics, differential settlement, strain com- patibility during temperature variations and loading, and so on. Scour at Outlet and Channel Degradation Local scour at the outlet of culverts based on discharge, cul- vert shape, soil type, duration of flow, culvert slope, culvert height above the bed, and tailwater depth can result in ser- vice life issues. These types of failures are avoided through proper design, such as following the guidance provided in FHWA Hydraulic Engineering Circular 14 (Figure 16). FIGURE 16 Example of outlet scour protection rock (Source: Ohio DOT).

18 Pipe–Headwall Connection Issues (Rotation, Settlement, Scour, etc.) Other types of drainage pipe system failures are difficult to predict and can include rotation, settlement or foundation fail- ure of headwalls, and scour at inlet and outlet ends (Figures 17 and 18). These types of failure are avoided through adequate subsurface investigation, appropriate headwall and scour pro- tection design, and good workmanship and materials. FIGURE 17 End treatment and scour failure at culvert outlet. FIGURE 18 Major culvert pipe system failure behind timber headwall. Gasket Degradation Within Pipe Joints Performance requirements for pipe joint gaskets are typi- cally based on short-term criteria; little is known about long- term degradation performance (Figure 19). FIGURE 19 Damage at joint in RCP revealing joint sealant as detected from video inspection.

Next: CHAPTER FOUR Pipe Materials »
Service Life of Culverts Get This Book
×
 Service Life of Culverts
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s National Cooperative Highway Research Program (NCHRP) Synthesis 474: Service Life of Culverts explores the time during which a culvert is expected to provide a desired function with a specified level of maintenance established at the design or retrofit stage.

This study is an update of NCHRP Synthesis 254: Service Life of Drainage Pipe (1998), which itself was an update of NCHRP Synthesis 50: Durability of Drainage Pipe (1978). Developments in plastic pipe, fiber-reinforced concrete pipe, polymeric-coated metal pipe, recycled materials, larger and more diverse structures, and sophisticated analytical soil-structure interaction modeling within the past 15 years led to the development of this report.

READ FREE ONLINE

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!