Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process (2021)

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.

6 Galvanizing steel structures is a large-scale industrial process in which steel components are dipped in molten zinc (∼840°F) to form a metallurgical bond between the steel part and a thick layer of zinc and zinc-iron intermetallic layers that protect the steel from corrosion. Galvanizing has been used in the United States on an industrial scale since the late 1800s, and the simplicity and success of the process as a corrosion protection scheme have led to more than 600,000 tons of zinc being used to galvanize steel elements in North America on an annual basis (American Galvanizers Association [AGA], 2019). As a process, hot-dip galvanizing (HDG) requires surface preparation of the steel part to be galvanized before it is dipped into molten zinc. This process involves immersing the steel part in a degreasing bath, followed by an acid pickling bath, and finally a fluxing bath, with rinsing between these process stops. After the steel part has been cleaned, pickled, and fluxed, it is submerged into a kettle of molten zinc that is held at a temperature of 830-850°F. During submersion, the steel component is lowered at an angle to allow for runoff of zinc when it is later extracted from the kettle. The steel part is allowed to reside in the molten zinc for a period of time sufficient to develop a metallurgically bonded zinc layer on the surface of the steel part; this dwell time varies with thickness/mass of the steel part being galvanized, but typically ranges from 5 to 20 minutes. Galvanizing facilities strive to minimize time in the kettle, since longer dwell times can result in unnecessarily thick coatings. When the coating has reached sufficient thickness, the steel part is extracted from the bath and then transferred to a cool- ing rack to cool at ambient temperature (in some cases, the galvanized part is quenched in a chromate solution). Incidences of cracking can occur in steel structures during galvanizing. Also, increased rates of cracking during HDG over the past two decades have been reported (Kinstler, 2005). Crack- ing associated with the HDG process has been shown to be a complex issue. Previous research on the mechanical properties of galvanized steel has shown that the mechanical properties of steel are generally not affected by HDG (Weigand & Nieth, 1964); this has contributed to the widespread use of galvanizing to protect steel against corrosion damage. However, it has been experimentally shown that the ductility of steel can decrease substantially when in contact with molten zinc (Kikuchi & Iezawa, 1982; Kinstler, 2005), a phenomenon commonly referred to as LMAC. LMAC occurs only in certain metal combinations that do not form intermetallic com- pounds (Kinstler, 2005). While a significant amount of research has been performed to explain the fundamental causes of LMAC in HDG, there is limited guidance on how to best control the occurrence of LMAC. In addition to LMAC, other factors and phenomena may contribute to cracking of welded highway structures due to the galvanizing process, such as geometric details of the structure, thermal stresses, hydrogen embrittlement, and strain-age embrittlement. C H A P T E R 2 Background and Literature Review

Background and Literature Review 7   A review of national and international standards and specifications, state department of transportation (DOT) practices (including materials selection, geometric detailing, welding procedures, and galvanizing procedures), galvanizers’ bath selections, and two case studies are provided in Appendix A. A summary of relevant literature is provided in Appendix B. Because LMAC (sometimes referred to as liquid metal embrittlement or LME) has shown to be directly related to the occurrence of cracking during galvanizing, a brief review of this phenomenon is presented. LMAC Mechanism LMAC is a phenomenon in which ductile metals experience substantial loss of tensile ductility in the presence of liquid metals. An example of LMAC is the embrittlement of brass by mercury (Huntington, 1912). The mechanism of LMAC is not well understood despite many proposed theories, such as increased air pressure in pre-existing cracks, the weakening of interatomic bonds by the presence of a liquid metal at the crack tip, and the formation of a weak-bonded alloy zone ahead of the crack tip (Nicholas & Old, 1979). LMAC is the result of the combination of three factors: (1) the presence of a net tensile stress (may be localized), (2) a specific liquid metal, and (3) a material condition or processing-induced sensitivity (Kinstler, 2005). The last two factors suggest that LMAC occurs only in certain metal combinations; typical examples of such combinations are steel-copper, stainless steel-zinc, aluminum-mercury, and copper-mercury (Kinstler, 2005; The Welding Institute, 2019). Many incidences of cracking that occurred during the steel galvanizing process have been attributed to LMAC. Microscopic examination of these cracks revealed concentration of zinc and other galvanizing bath additives along the crack paths and at the crack tips (Kinstler, 2005; Sedlacek et al., 2004); this is illustrated in a case study presented in Appendix A. Katzung and Schulz (2005) also concluded, based on experience and comprehensive metallographic experi- ments, that most cracking damage related to HDG can be traced back to LMAC. Kinstler (2005) reported that instances of LMAC occurring between steel and liquid zinc were observed, even though the formation of intermetallic compounds is the basis for HDG and metal combinations that form intermetallic compounds have been believed to not cause LMAC. Researchers have studied the effect of molten zinc on the reduction of steel ductility. Kikuchi and Iezawa (1981) reported a sudden drop in ductility of steel tension specimens when immersed in molten zinc but no effect on ductility when tested in air at the same elevated temperature (ET), as shown in Figure 2-1. This behavior was later reproduced by Kinstler (1991), as shown in Figure 2-2. More recently, Beal et al. (2012) investigated the embrittlement of zinc-coated (electro galvanized) high-manganese steel after treatment with a heat-affected zone (HAZ) simu- lator. A reduction of ductility (i.e., embrittlement) was observed within a limited temperature range under certain strain rates due to liquid zinc’s ET. No embrittlement was observed for steel under very low strain rates. Because the hot tensile tests under high strain rates lasted only a few seconds, it appears that intermetallic compounds did not have time to form to sup- press embrittlement, but such compounds formed during the tests under low strain rates. Thus, for LMAC to occur, a critical stress level has to be reached in order for the liquid zinc to be consumed by the formation of intermetallic compounds. LMAC, as measured by the specimen’s strain at failure, was used in this research as an indica- tor of susceptibility of steel specimens to cracking during galvanizing, and the strain at failure was used as the criterion for assessing the level of susceptibility to LMAC. In addition, this study considered the hardness value of 270 HV (Vickers hardness) suggested by prior research (Elboujdaini et al., 2004) as an indicator of susceptibility to LMAC, and it was used in evaluating the data obtained from hardness measurements.

8 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process 0 20 40 60 0 5 10 15 20 σ( kg f/ m m 2) δ(mm) SMA50A Stress Displacement Curve Room Temperature after Zn coating (10% sulfuric acid solution) Room Temperature (in the air) 460°C in air (10% sulfuric acid solution) 460°C in air (grinding treatment) 460°C in Zn (10% sulfuric acid solution) 460°C in Zn (grinding treatment) Figure 2-1. Stress/Strain curve at room temperature (RT), } 460çC, and with acid pickling or abrasive cleaning (grinding), with and without zinc, adapted from Kikuchi & Iezawa (1981). St re ss Strain Without Zinc: UTS=58.7-60.1 ksi Yield=33.7-35.2 ksi Elong=23.6-25.8% With Zinc: UTS=55.6-56.5 ksi Yield=32.1-33.6 ksi Elong=5.0% (Curves offset for clarity) Figure 2-2. Stress/Strain curve with and without zinc at 840çF, adapted from Kinstler (1991).

Background and Literature Review 9   Although the reduction of ductility is the main result of LMAC, the susceptibility of structural steel to cracking during the galvanizing process is influenced by a number of factors. There are multiple combinations of factors that would increase the likelihood of LMAC, as discussed in this report: • Localized Stress – Residual stresses arising from rolling, cold-working, welding, and cutting processes – Imperfections resulting from thermal cutting – Stress concentration factors (SCFs) related to the geometric shape of the structural com- ponents and connections – Induced stresses due to the thermal cycle of the galvanizing process • Environment – Chemical composition of the galvanizing bath • Time – Dwell time in the galvanizing kettle – Cooling period after galvanizing • Condition – Chemical composition and thermal/mechanical history of the steel and strength and mechanical properties of the steel This concept is illustrated in Figure 2-3, adapted from Kinstler (2005). Figure 2-3. Confluence of factors that contribute to the occurrence of LMAC, adapted from Kinstler (2005).