壓水式反應器二次側結構組件腐蝕劣化的防蝕研究

Corrosion Mitigation of Structural Materials in the Secondary Cooling System of Pressurized Water Reactors

賴煒升(Wei-Sheng Lai)

指導教授 ： 葉宗洸

國立清華大學/原子科學院/工程與系統科學系/碩士(2020年)

摘要

壓水式反應器二次側結構組件的腐蝕情況以蒸氣產生器最為嚴重，因為在蒸氣產生器內的管件與支撐用的管板之間會有縫隙產生，容易導致間隙腐蝕。此外，腐蝕產物堆積在縫隙硬化後會擠壓管件造成凹陷，在管件U型彎曲部分存在張應力，容易出現應力腐蝕龜裂。
目前核三廠使用聯胺(Hydrazine)和乙醇胺(Ethanolamine, ETA)作為水質控制的添加劑，蒸汽產生器的酸鹼值約控制在9.6，聯胺添加濃度約為30 parts per billion (ppb)，乙醇胺濃度約為1 ppb，其他雜質濃度都維持在近偵測極限。此外，聯胺會間接的影響導電度，因為聯胺在高溫下會反應分解成氨，導電度主要的貢獻來自於溶於水中的氨，電廠內測量值約為9μS/cm。陽離子導電度約為0.2μS/cm。
應用於壓水式反應器二次側組件的材料，主要是鎳基合金、奧斯田鐵不銹鋼、低合金鋼以及銅合金等，而鎳基合金包含600MA、600SR、600TT、800NG以及690TT等。抑制二次側組件的腐蝕程度以減少腐蝕產物的產出與傳輸，必須要能透過控制系統中的pH值、減低或控制系統中的氧化劑濃度、控制系統中的導電度與雜質來達成。
因此，實驗設計為透過壓水式反應器二次側水循環模擬迴路，針對600合金和690合金試片進行均勻腐蝕試驗、U-bend應力腐蝕試驗、間隙腐蝕試驗和CBB (Crevice Bent Beam)腐蝕試驗。U-bend應力腐蝕試驗作法為將試片彎曲成U型並固定於夾具內，試片於夾具內的曲率為固定12.5%之應變量。間隙腐蝕試驗作法為利用刻有溝槽的耐高溫絕緣材料夾住試片，利用溝槽縫隙來形成間隙腐蝕條件。CBB腐蝕試驗使用夾具彎折試片並維持其應變量在1.2%，並在試片上方鋪一層親水碳布來形成應力與間隙條件。水化學條件則是透過注入不同濃度的氫氧化銨調整系統中的導電度與pH值，並通入氮氣使得水環境控制在低溶氧狀態。測試後的試片則利用掃描式電子顯微鏡（Scanning Electron Microscope, SEM）與能量散佈式X光頻譜(Energy Dispersive X-ray Spectrometer, EDX) EDX）對試片表面產生的裂縫的分布情形及裂縫擴展情形進行分析，探討在 280 oC、8.3MPa條件下運轉時，不同的水化學環境和不同試片條件對於600合金和690合金試片的影響。
實驗發現600合金和690合金氧化物的結構在純水條件下以氧化鎳(NiO)為主和少數氧化鉻(Cr2O3)、氧化鎳鉻(NiCr2O4)。在pH=9.6的酸鹼值下，不論是否添加聯胺，表面氧化層為尖晶石結構的鎳鐵氧化物(NiFe2O4)。
均勻腐蝕試驗中在pH=9.6觀察到沿晶浸蝕的現象，由其歷經退火(Mill Annealed, MA)的試片更加明顯，聯胺的添加可以抑制沿晶浸蝕。質量變化試驗中發現純水條件試片增重最多，添加聯胺條件的增重減少許多。
間隙腐蝕試驗在純水條件因為導電度不足觀察不到間隙腐蝕，提升導電度後除了氧化物在間隙更深處生成之外，間隙內部可以觀察到孔蝕與氮化鈦析出物誘發的孔蝕起始。
CBB試驗中即使含有應力與間隙，600合金和690合金沒有觀察到孔蝕與裂縫，但是氧化物生成速度比平板試驗還要快速。表示1.2 %應變量提供的應力仍不足以使試片產生裂縫，但仍須考慮氧化膜的保護效果與裂縫是否被氧化膜遮蔽。聯胺的添加能抑制氧化層在CBB條件的生長。
應力腐蝕龜裂試驗以定量分析與局部觀察來探討環境對於600合金和690合金的影響。定量分析結果顯示在純水與pH=9.6的條件下，長度0到10微米的裂縫數目增加趨勢相似。在大於10微米的裂縫中，除了600合金在pH=9.6的條件下，裂縫長度10到20微米的裂縫數目是持續成長的之外，裂縫數目的增長都是趨緩的。聯胺的添加則明顯抑制裂縫的生長，可以觀察到在所有長度區間的裂縫成長數目都大幅降低。此外，可以觀察到600合金在起始裂縫數目都比690合金少，經過電解觀察表面證實690合金試片表面含有的氮化鈦析出物遠大於600合金，堅硬的析出物在試片彎折時容易成為差排堆積處，造成裂縫的起始。裂縫的局部觀察發現裂縫在經過1500小時的試驗後可以觀察到小裂縫的合併與裂縫寬度的增加，這個現象發生在所有條件中，即使添加聯胺仍在 600合金觀察到裂縫變寬。
本實驗在均勻腐蝕試驗觀察到軋制退火熱處理的試片出現沿晶浸蝕現象，而固溶熱處理的沿晶浸蝕則不明顯。因此我們認為固溶熱處理有更優異的材料微結構，但在其他條件的試驗中，不同熱處理並無觀察到明顯差異。在應力腐蝕龜裂的定量分析中，690合金對抗裂縫擴展能力比600合金更優異。所以在本試驗中690SA合金是最佳的材料。

關鍵字

壓水式反應器二次側 ； 600/690合金 ； 應力腐蝕龜裂 ； 間隙腐蝕 ； 聯胺

並列摘要

Steam generators show the most severe condition of the corrosion in PWR secondary side because there are crevices between the tube bundles and support tube sheet, it may cause crevice corrosion in the intermediate gap. The accumulated sludge in the gap will harden gradually, it will press the tubes to cause denting. Besides, the inner rows of the tube bundles at the U-bend portion have larger tensile stress, these spots will cause stress corrosion cracking easily.
Currently, Maanshan Nuclear Power Plant utilizes hydrazine and ethanolamine to control water quality. The pH value in steam generator blow down is around 9.6 at 25 oC. The concentration of hydrazine and ethanolamine are 30 ppb and 1 ppb respectively. Other impurity concentrations are quite low to near detection limit. In addition, hydrazine has the impact on the conductivity in secondary side because hydrazine decomposes into amine at high temperature which mainly contributes the conductivity in secondary side. The conductivity is around 9 μS/cm and cation conductivity is around 0.2 μS/cm in in-situ measurement.
For the secondary side components, the main materials are nickel-based alloy, stainless steel, low alloy steel and copper alloy. In order to improve reliability and to reduce corrosion products from transporting from secondary side to steam generator, the pH value, oxidant concentration, conductivity and impurities need to be controlled.
Therefore, the designation of the experiment will test Inconel 600 alloy and Inconel 690 alloy in simulated PWR secondary side water chemistry circuit through different sample conditions and different water chemistry conditions. Uniform corrosion test was conducted by flat specimen method. Stress corrosion cracking test was conducted by U-bend specimen method. Crevice corrosion test was conducted by multiple assembly crevice method. Combined stress and crevice corrosion test was conducted by crevice bent beam method. In U-bend method, specimens are bent into a U-shape. The curvature of the specimens is in a fixed value of 12.5% strain. In crevice corrosion method, the specimen is clamped between two fixtures which have high temperature resistance and electrical insulation with grooves around it to create the crevice corrosion condition. In crevice bent beam method, the specimens are bent by the fixtures with the strain of 1.2 %. A layer of hydrophilic carbon cloth is placed on the top of the specimen to form crevice condition. Then, different concentrations of ammonia hydroxide were injected to control the conductivity and pH value. The dissolved oxygen in water is controlled to 300 ppb by nitrogen. The specimens were tested under 280 oC and 8.3 MPa. After the test, the distribution of the cracks generated on the surface of the specimens and the propagation of the cracks were analyzed by scanning electron microscope (SEM). We analyze the composition of the oxide layer on the surface by energy-dispersive X-ray spectroscopy (EDS) and laser Raman spectrometry (LRS) to find out the influence of ammonia hydroxide addition and different sample conditions.
The results showed that the structures of oxide on alloy 600 and alloy 690 are mainly nickel oxide in pure water condition. There are also some chromium oxide and nickel chromium oxide (NiCr2O4). The oxide structure becomes nickel iron oxide in pH value of 9.6.
We found etching along the grain boundary in the sample with mill-annealed heat treatment in pH value of 9.6 in the. It can be mitigated by the addition of hydrazine. The mass gain analysis for uniform corrosion test showed that the specimens in pure water condition had larger mass gain, instead those in hydrazine addition had less mass gain.
We can’t find pits and cracks in crevice corrosion test in pure water condition due to the low conductivity. After increasing the conductivity to 10 μS/cm, there are some pits formed at the inner part of crevice. The oxides also formed in deeper part of crevice.
In the CBB tests, although combination of the stress and crevice condition, we can’t find any pits or cracks. It indicated that the stress is not enough to cause degradation. However, we need to consider the pits may be covered by oxide film or the matrix may be protected by oxide film. Besides, the formation of oxide film is faster than uniform corrosion test in the same water chemistry condition. Hydrazine can mitigate the formation rate of oxide film.
In stress corrosion cracking tests, the quantitative analysis showed that there are similar trends of crack growth rate for crack length lower than in pure water condition and pH value of 9.6 condition. The crack growth rate tended to alleviate for cracks over than 10 μm except for alloy 600 in pH value of 9.6 condition. Hydrazine addition effectively mitigated the crack growth for all specimens. In addition, we need to be aware of the influences of TiN inclusion on the crack initiation of alloy 690.
In uniform corrosion tests, we observed etching along the grain boundary on the specimens with mill-annealed heat treatment, but not on the specimens with solution-annealed heat treatment. As a consequence, we believed the materials with solution-annealed treatment have better microstructure against etching along the grain boundary. However, no obvious differences were observed in others tests between the heat treatments. In the quantitative analysis of SCC tests, alloy 690 showed better resistance for crack propagation. According to the results, we believed alloy 690SA has the best performance among the selected materials.

並列關鍵字

PWR secondary side ； Inconel 600/690 ； stress corrosion cracking ； crevice corrosion ； hydrazine

參考文獻

[1] P. L. Andresen, “Emerging issues and fundamental processes in environmental cracking in hot water”, Corrosion, vol. 64, 439-464, 2008

[2] Nishimura, R., & Maeda, Y. “Metal dissolution and maximum stress during SCC process of ferritic (type 430) and austenitic (type 304 and type 316) stainless steels in acidic chloride solutions under constant applied stress” Corrosion Science, vol. 46, pp. 755–768, 2004

[3]H. H. Uhlig, Physical metallurgy of stress corrosion fracture, InterScience Publishers, edited by T.N. Rhodin, Pittsburgh, Pennsylvania ,1959

[4] Lynch, S. P. “Mechanistic and fractographic aspects of stress-corrosion cracking (SCC)”, Corrosion reviews, vol. 30, pp. 3–89, 2012

[5]H. L. Logon, “Film-rupture mechanism of stress corrosion,” Journal of Research of the National Bureau of Standards, vol. 48, pp. 99-105, 1952