| 摘要: | 本研究探討反應爐壓力槽(Reactor Pressure Vessel, RPV) SA533B1鋼材之疲勞裂縫起始及成長機構。測試材料包括四種不同硫含量(0.0062 wt% ~ 0.035 wt%)之SA533B1鋼材，進行室溫及高溫下之高週、低週疲勞及疲勞裂縫成長速率測試。疲勞裂縫長度分別以交流電位降法及開口位移計量測，以比較不同量測技術之差異，並以應變片量測疲勞裂縫成長之裂縫閉合程度，瞭解裂縫閉合現象對疲勞裂縫成長之影響。同時，藉掃瞄式電子顯微鏡(SEM)及穿透式電子顯微鏡(TEM)觀察鋼材經疲勞測試後之微觀組織變化，以探討疲勞裂縫起始及成長模式，提供RPV鋼材疲勞壽命評估之依據。 實驗結果顯示，硫含量高低對室溫SA533B1鋼材高週疲勞限無顯著影響，疲勞限約為650 MPa。高溫(300?C)高週疲勞限及疲勞壽命較室溫者略高，尤以硫含量低者（0.006 wt% S）疲勞限最高，達702 MPa。低週疲勞測試方面，室溫及高溫之鋼材低週疲勞壽命與硫含量高低無明顯關聯。低週疲勞壽命可藉應變－壽命經驗式及SWT關係式預估之。另外，鋼材中之Al、S、Ca、Mn及O等元素形成之夾雜物(Inclusion)對高週疲勞壽命有顯著影響；但低週疲勞壽命較不受夾雜物影響。高週疲勞裂縫常起始於夾雜物處，300℃時裂縫甚至起始於試片內部夾雜物處。此外，SA533B1鋼材於300℃發生之動態應變時效及伴隨高溫塑性變形之晶粒細化效應有助於鋼材強度提升，對高週及低週疲勞壽命均有助益。但相反地，高溫塑性變形促使強迫固溶於基地的碳化物析出，造成鋼材強度下降，對高週及低週疲勞壽命卻有損害。晶粒細化與碳化物析出兩機制彼此競爭結果，晶粒細化效應較顯著。因此，300?C高週與低週之抗疲勞性較室溫略佳，且經300?C疲勞測試鋼材之室溫硬度值明顯呈上升現象。 疲勞裂縫成長方面，藉交流電位降量測法可準確量測高溫疲勞裂縫長度。研究結果發現，不同硫含量鋼材疲勞裂縫成長速率均可由Paris Law表示成da/dN (mm/cycle) = 1.98×10-8×(?K)2.54 (MPa )；和硫含量及測試溫度(室溫、150℃及300oC)無明顯關係，但當硫化物方向與疲勞裂縫成長方向平行時，裂縫成長速率較快。以應變片量測裂縫閉合發現，應力比下降至R = 0.13之受力情況，才出現裂縫閉合現象，因此，R = 0.2之裂縫成長速率不需修正。另外，裂縫閉合程度與負荷歷程及裂縫尖端塑性區大小有密切關係，裂縫閉合程度隨著裂縫尖端塑性區重疊而趨嚴重。 穿透式電子顯微鏡分析結果證明，疲勞累積傷害與次晶粒間之方位差無明確關聯性，而高溫高週及低週疲勞測試後晶粒寬度變小，碳化物沿著晶界、次晶界析出，與SEM觀察結果一致。 綜合而言，SA533B1鋼材雜質含量對高週疲勞壽命有顯著影響，若電廠建廠之初能提高鋼材純度，減少雜質含量，將有助於管嘴附近高週疲勞壽命提昇。由本研究結果得知，高週疲勞與低週疲勞之裂縫起始機構不盡相同，故以低週疲勞數據之應變振幅乘以楊氏模數轉換而得之S-N曲線，無法精確預測高週疲勞S-N曲線。另外，當雜質方向與疲勞裂縫生長方向平行時，裂縫生長會加速，電廠尤需注意監控。 High-cycle, low-cycle fatigue tests and fatigue crack propagation measurements were conducted on SA533B1 steels, reactor pressure vessel (RPV) steels, with four levels of sulfur content (0.0062 wt%~ 0.035 wt%) at room temperature and 300 ℃. Efforts were made to study the effects of sulfur content on the fatigue degradation of RPV steels. Fractographic and microstructural examinations by scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were performed to investigate the fatigue crack initiation and propagation mechanism. To make a comparison of different crack measurement techniques, fatigue crack length measurements were concurrently taken on the same specimen by an alternating current potential drop technique and a crack opening displcement gauge. Fatigue crack opening levels were also measured to characterize the fatigue crack closure effect on fatigue crack growth behavior. The results of high-cycle fatigue (HCF) limit were observed to be ~650 MPa, around the yield strength of SA533B1 steel, and showed little or no dependence on sulfur content at room temperature. At 300℃, the fatigue limits were slightly higher than those tested at room temperature, notably for the steel with a sulfur content 0.006 wt% whose fatigue limit at 300℃ was up to 702 MPa. The low-cycle fatigue (LCF) limit showed little or no dependence on sulfur content, but significantly depended on the test temperature. Fatigue life could be predicted either by means of the strain-life equation or the SWT (Smith, Watson, and Topper) parameter for specimens under low-cycle fatigue conditions. The strain-life equation for SA533B1 steel in air was irrespective of the sulfur content, but dependent on the test temperature. HCF life was significantly affected by the inclusions which are composed of O, Al, S, Ca, and Mn. The inclusions were randomly distributed on the surface or in the interior of the specimen. Fractographic examination results suggested that inclusions near the specimen surface served as the crack initiation site for a majority of the fatigued specimens tested at room temperature. For those tested at 300℃, some cracks were identified to initiate around the inclusions in the interior of the fatigued specimens. However, the effects of inclusion on the LCF are less significant than HCF. No LCF specimens were observed to initiate at/around the inclusions. The fatigue resistance of SA533B1 steel subjected to HCF and LCF tests at 300℃ was improved by the combined effects of dynamic strain aging (DSA) and grain size reduction. DSA and grain size reduction increase the steel strength, accordingly improve the HCF and LCF limits at 300℃. But, concurrently, the precipitation of carbides/nitrides in SA533B1 steel leads to a decrease in the steel strength. The grain size reduction and the precipitation compete with each other. The grain size reduction is dominant and the net effect is reflected in an increase in hardness value. The crack growth rate measurements for steels of different sulfur contents tested at room temperature and 300℃ can be characterized by the same material constants C = 1.98*10-8 and m = 2.54 for Paris law, where da/dN is in mm/cycle and ?K in MPa . It implies that fatigue crack growth rate has little or no dependence on sulfur content and test temperatures (25℃, 150℃ and 300℃). But fatigue crack growth rate becomes faster, when crack propagation runs parallel to the sulfide orientation. Fatigue crack closure became evident till the stress ratio (R value) was decreased to the values smaller than 0.13. Thus, for the tests at the stress ratio of R = 0.2, it is not necessary to calibrate the ?K value. The results of crack opening measurement show that the levels of crack closure significantly depend on the loading history and plastic zone size ahead of crack tip. The phenomenon of crack closure was observed to be more significant for the specimens with a through crack than those with a surface crack. Smaller subgrains and a greater number of precipitates were identified with the specimens tested at 300?C by SEM and TEM examinations. No clear relationship was found between fatigue damage and the mis-orientation changes of cell walls (or subgrain boundaries) in the fatigued samples of SA533B1 steel. |
