2018 年 59 巻 7 号 p. 1124-1129
Secondary ion mass spectroscopy (SIMS) was used to examine the hydrogen atoms in low-carbon boron (10B21) steel screws. The effects of baking and tempering treatments on the hydrogen-induced embrittlement (HIE) susceptibility of the screws were investigated. SIMS results confirmed that hydrogen concentration decreased with increasing baking duration, and thus increased the engineering reliability of the screws. For low-temperature applications, 10B21 screws must be baked for a longer duration to prevent HIE. The observed tempered martensite was composed of ferrite and cementite, which could limit the movement of hydrogen atoms. At higher tempering temperature, the structure of the screw matrix became finer, reducing the HIE susceptibility. 10B21 screws tempered at a high temperature thus had good ability to resist low-temperature HIE.
Fig. 11 HIE susceptibility of 10B21 screws as function of tempering temperature and baking duration.
Low-carbon steel is widely used in the production of fasteners. Hydrogen-induced embrittlement (HIE) may occur as a result of the introduction of hydrogen into the steel during zinc plating processes.1) Proposed methods for preventing HIE include baking, tempering, and adding alloy elements.2,3) HIE susceptibility of steel decreases with increasing tempering temperature, which is attributed to a reduction in the number of dislocations and hydrogen traps.2) Adding particular alloy elements increases grain boundary cohesion and thus decreases intercrystalline embrittlement.3)
Screws are commonly treated using quenching and tempering to enhance their toughness to reduce HIE susceptibility.4) The nose of the time-temperature-transformation (TTT) diagram for low-carbon steel is very close to the left that is not easy to produce the martensite transformation.5) Adding Ni, Cr, or Mo into low-carbon steel leads to the formation of martensite, which enhances hardenability.6) Ghali et al.7) found that adding boron (B) can effectively enhance the hardenability of carbon steel. Low-carbon boron (10B21) steel is most commonly used to fabricate screws in industry due to its relatively low cost. In general, HIE does not occur easily in the low-carbon steel fasteners. The hydrogen-induced cracking of 10B21 screws in wooden structures was described in our previous study.8) In the present study, the effects of tempering microstructure of 10B21 steel screw on the HIE susceptibility are investigated.
Many studies have been conducted on the role of the microstructures of 10B21 steel in resisting the harmful effects of hydrogen.9) Martensite is most susceptible to hydrogen damage, whereas temperted martenstie is least susceptible.10) A threaded fastener specimen can essentially be examined with several notches that act as stress risers under hydrogen embrittlement (HE) test, which usually fractures at the thread root radius, shank, or the head-to-shank fillet. Notches sensitivity produced by threads on fasteners has an effect on the maximum load and affected by mechanical properties of the 10B21 screw. The effect of hydrogen on 10B21 steel depends on the tempering conditions and the amount of hydrogen present. Hydrogen-induced cracking is often related to the tempering temperature. Many studies have shown that deformation resistance is not the dominant factor in HIE susceptibility and that microstructural features may play a more important role in HIE resistance.9,11)
The ductility of 10B21 steel sharply decreases at low temperature (−20 to −220°C); there is an obvious ductile to brittle-transition-temperature (DBTT).12) Microstructural features have a large effect on DBTT, and may thus affect low-temperature HIE susceptibility. The occurrence of HIE is directly related to the diffusion rate of hydrogen,13) even though this rate tends to decrease with decreasing temperature, implying reduced the low-temperature HIE susceptibility. Jackson et al.14) found that the HIE resistance of 304L austenitic stainless steel at −50°C was worse than that at room temperature. This result was attributed to the more rapid growth of cracks in steel at low temperature. The low-temperature HIE susceptibility of 10B21 steel has not been thoroughly investigated.
The present study detected hydrogen in 10B21 screws. The effects of baking duration after the plating process on the engineering reliability of screws were examined via statistical analysis. Based on previously reported experimental results, the screws were tempered at temperatures of up to 360°C to investigate the effects of structural morphology and tensile deformation resistance on the low-temperature HIE susceptibility of 10B21 screws.
Low-carbon boron (10B21) steel was used to fabricate wooden building screws. The chemical composition of the steel is summarized in Table 1. 10B21 steel wire rod (Φ = 5.5 mm) was extruded after pickling with zinc phosphate solution, and then screws were formed using cold forging. The screws were austenitized at 950°C for 30 min in an air furnace, and then quenched in water. Then, the quenched screws were tempered (at 220, 260, and 320°C) for 30 or 60 min. The tempered screws were then galvanized. According to heat treatment conditions, the 10B21 screws are designated as Q950, QT220-30, QT260-30, QT260-60, QT320-30, and QT320-60, respectively. Finally, the screws were baked at 215°C for 0–16 h.
Tensile strength results for 10B21 steel specimen obtained at various tempered conditions are shown in Fig. 1. The as-received specimen has spheroidal cementite; it has the lowest strength and highest ductility. Tensile strength decreased and the ductility increased with increasing tempering temperature. The effect of tempering duration on the tensile properties was not obvious due to the size effect. The structure of small size screw (Φ < 5.5 mm) has been sufficiently phase-transformed after tempering for 30 min. The tensile properties of 10B21 screw specimen were not obvious change even through tempered for 60 min. To reduce the HIE susceptibility, the 10B21 steel specimens subjected to by tempering treatment. However, according to tensile test results (HRC > 32; UTS > 1250 MPa), the 10B21 steel specimens still had high HIE susceptibility.15) Therefore, the HE tests for 10B21 screws were further conducted.
Tensile properties of 10B21 steel specimens obtained under various tempering conditions.
Twist-off strength and HE (room temperature, −15°C, and −30°C) tests were performed in the atmosphere to estimate the HIE susceptibility of 10B21 screws (Fig. 2). The screws were fastened in increments of 0.5 N·m until fracture to acquire the twist-off strength. For HE testing, according to ANSI/ASME B 18.6.4, screws subjected to a pre-stress of 80% (22 N·m) of the twist-off strength were fastened to a steel plate, and held for 24 h. The microstructures of the steel specimens were observed using optical microscopy (OLYMPUS, Tokyo, Japan). The fracture characteristics of the screws were examined using field-emission scanning electron microscopy (HITACHI, Tokyo, Japan). To examine the effect of baking on dehydrogenation, the transverse cross-section of the screws was examined by secondary ion mass spectroscopy (SIMS) using a CAMECA IMS-6f analyzer (CAMECA, Genevilliers, France) with Cs primary ion beam. Note that the screw specimens were placed in the analysis chamber about 24 h before detection and then cooled using liquid nitrogen to minimize hydrogen background. Based on the fracture torque measurements of screws baked for various durations, Weibull analysis was performed to evaluate the engineering reliability. In addition, the crystalline phases of the 10B21 screws we re examined using X-ray diffraction (XRD; Rigaku, Tokyo, Japan).
Schematic diagram of twist-off strength and HE testing.
The macrostructure characteristics and the fracture surface of baked (8 h) QT220-30 screw after HE testing are shown in Fig. 3. This screw fractured in the threaded zone (Fig. 3(a)). The periphery of the fractural surface (Fig. 3(b)) indicates brittle fracture (Fig. 3(c)), which is attributed to hydrogen-induced intergranular fracture.16,17) In the central part, there are many dimples, created by tensile stress, indicating typical ductile fracture (Fig. 3(d)). The peripheral structure of the screw near the thread is the HIE susceptible region; the central structure can resist HIE. In this case, HIE occurred at the peripheral structure of the screws and stress concentrated in the screw threads, resulting in ductile fracture in the central part of the screws.
(a) Photograph, (b) macrostructural characteristics and (c, d) fracture surfaces of HIE screw (QT260-30).
The sources of hydrogen in the screws are the pickling and plating processes. Hydrogen content in the screws was detected using SIMS to clarify the effect of baking duration on hydrogen release (Fig. 4). The mass spectrum for hydrogen was measured at the transverse cross-section periphery of the screws. The absolute hydrogen content in the screws is not provided due to the lack of a standard sample for hydrogen; only the relative changes of hydrogen content with baking duration are presented. The integral area under the curves decreases with increasing baking duration, indicating that hydrogen content in the 10B21 screws significantly decreased. Compared with unbaked screws, the proportion of hydrogen content in the screws respectively decreased by 74% and 82% after baking for 8 and 16 h, respectively, and thus affected HAE susceptibility of screw.10) In order to prevent HIE, the baking duration should exceed 8 h. The effect of baking duration on the ultimate torque strength determined using Weibull analysis18) is plotted in Fig. 5. The data distribution of the screws becomes more concentrated and the peak increases with increasing baking duration. As the peak shape becomes more obvious, the peak moves toward the characteristic value. The data distribution range is narrow, indicating that it is easy to predict when a screw will fail. Although the ultimate torque strength of the screws slightly decreased after baking for a long duration, the reliability of the screws increased.
High mass resolution analyses of QT220-30 screws obtained using various baking durations.
Failure probability density function curves of baked QT220-30 screws.
The baked (8 or 16 h) QT220-30 screws are subjected to low-temperature (−15°C) HE testing; the results are shown in Fig. 6. The resistance to low-temperature HIE susceptibility of the screw baked for 8 h was poor. The slip system of the screw material decreased, rapidly deteriorating resistance to crack growth at low temperature. Note that the screws baked for 16 h did not exhibit HIE. Although the microstructure of these screws was not transformed, their HIE susceptibility was lower than that of the screws baked for 8 h (Fig. 4).10) Screws tempered at lower temperature must be baked longer to avoid HIE at low temperature.
Low-temperature HE testing results of QT220-30 screws after baking for various durations (×3: 3 brittle fracture screws).
Excessive baking time is unsuitable for industrial applications due to cost. Therefore, the screws were tempered at higher temperatures to investigate the effects of tempering microstructure on their mechanical properties and HIE. Figure 7(a) shows the twist-off strength of the 10B21 screws produced with various tempering temperatures after baking for 8 h. The average fracture torque (hollow cross) of the screws decreased with increasing tempering temperature, which is associated with some martensite transforming into cementite and ferrite.19) For tempered effect on HE testing (Fig. 7(b)), the QT220-30 screws have serious HIE, which is associated with the quenched structure not completely transforming into tempered martensite. The QT260-30 and QT320-30 screw specimens rarely exhibited HIE, which is attributed to the fine tempered martensite structures effectively capturing the hydrogen atoms and thus preventing hydrogen accumulation.20)
(a) Twist-off (cross: average fracture torque; ×3: 3 screws specimens, and so on) and (b) HE testing results of 10B21 screws for various tempering temperatures (×4: 4 brittle fracture screws).
To investigate the effects of tempered microstructure on HIE susceptibility, the microstructures of 10B21 screws obtained at various tempering temperatures were examined (Fig. 8). Figures 8(a) and (b) show tempered martensite with ferrite and cementite distributed in the QT260-30 screw matrix. The microstructure did not obviously change even after tempering for 60 min. The morphology images of QT320 screws (Figs. 8(c) and (d)) show finer tempered martensite structures similar to the lamellar pearlite structure, which composed of ferrite (the light gray regions) and cementite (the black particles). Higher tempering temperature led to increases in the amounts of ferrite and cementite, which reduced the mechanical strength of the screws. The cementite acted as hydrogen trapping sites, which can limit the movement of hydrogen atoms and thus prevent HIE.21) The tempered 10B21 screws are examined using XRD analysis; the results are shown in Fig. 9. The peaks at 51° and 74° are not obvious, indicating that there was little retained austenite in the steel. This result is attributed to most of carbon directly contributed to low-carbon martensite, so there was not enough carbon to form retained austenite.22) The three main peaks correspond to the contributions of ferrite and martensite. The intensities of these peaks increased with increasing tempering temperature; this is beneficial for preventing HIE, which results from cementite.
Microstructure metallographs of 10B21 screw specimens obtained under various tempering conditions.
XRD patterns of tempered 10B21 screws.
10B21 screws treated at higher tempering temperature were subjected to low-temperature HE testing; the results are shown in Fig. 10. HIE did not occur for QT260-30 and QT320-30 screws at −15°C. For QT320-30 screws, HIE did not occur even at −30°C. Higher-temperature tempering made the screw structure finer, allowing it to effectively trap hydrogen atoms at grain boundaries. The cementite provided hydrogen trapping sites, blocking the movement of hydrogen atoms.23) High-temperature tempered microstructure was a great contribution to HIE at the low-temperature environment, and significantly reduced the cost of baking process.
HE results of QT260-30 and QT320-30 screws for tests conducted at −15°C and −30°C.
The effects of tempering temperature and baking duration on the HIE susceptibility of 10B21 screws are summarized in Fig. 11. Baking treatment decreased hydrogen content in 10B21 screws, which decreased HAE susceptibility and increased engineering reliability. A small increase in tempering temperature resulted in changes in the microstructural features, which reduced the HIE susceptibility. The high-temperature tempered microstructure had good resistance to HIE even at low temperature.
HIE susceptibility of 10B21 screws as function of tempering temperature and baking duration.
This study investigated the effects of baking duration and tempering temperature on the low-temperature HIE susceptibility of 10B21 screws. Hydrogen content in baked screws could be significantly decreased and engineering reliability should be increased. Screws tempered at lower temperature must be baked longer to avoid HIE at low temperature. With higher-temperature tempering, the torque strength of 10B21 screws decreased, decreasing HIE susceptibility. Based on microstructural examinations, an increase in tempering temperature made the microstructural features finer. The cementite in the tempered microstructure acts as hydrogen trapping sites. 10B21 screws treated at higher tempering temperature had better low-temperature HIE resistance, which resulted from their finer structure and the contribution of cementite.
The authors acknowledge Dr. Kuan-Jen Chen for assistance in technical services (SIMS) by Ministry of Science and Technology (MOST) Instrument Center at National Cheng Kung University (NCKU) and MOST, Taiwan for financially supporting this study under Grant No. MOST 105-2628-E-006-001-MY2.
