Corrosion of iron stainless steels in molten nitrate salt

被引:89
作者
Kruizenga, Alan [1 ]
Gill, David [2 ]
机构
[1] Sandia Natl Labs, POB 969, Livermore, CA 94551 USA
[2] Sandia Natl Labs, Albuquerque, NM 87185 USA
来源
PROCEEDINGS OF THE SOLARPACES 2013 INTERNATIONAL CONFERENCE | 2014年 / 49卷
关键词
Austeneitic Steel; Molten Nitrate Salt; High Temperature Corrosion;
D O I
10.1016/j.egypro.2014.03.095
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
Energy storage for concentrating solar power (CSP) is a major area of research that seeks to lower the levelized cost of electricity within the aggressive SunShot goals of 6 / kW-hrth[1-3]. One viable approach is sensible thermal energy storage (TES), which currently utilizes molten nitrate binary salt, stored at 575 C in the hot tank of a two tank system [4, 5]. Increasing the temperature limit within the hot tank requires a detailed understanding of materials corrosion behavior, in addition to salt thermal stability properties. High temperature nickel based alloys are the logical choice for strength and corrosion resistance as elevated temperatures will increase corrosion kinetics, however, the cost of nickel based alloys are nearly four times more expensive than iron based steels [6]. For this reason iron based stainless steels, specifically 321SS and 347SS (nominally Fe-17Cr-9Ni), were chosen for investigation at several temperatures in nitrate salt. 316SS, an elementally similar alloy, was susceptible to stress corrosion cracking while tested at Solar Two [4]. It was suggested that alloys with stabilizing additions of niobium (347SS) or titanium (321SS) would mitigate this deleterious behavior. Flat coupon samples were immersed in binary nitrate salts at temperatures of 400, 500, 600, and 680 degrees C, with air sparging on all tests. Samples were nominally removed at intervals of 500, 1000, 2000, and 3000 hours to acquire data on time varying weight gain information while simultaneously employing metallography to identify corrosion mechanisms occurring within the melt. Corrosion rates varied dramatically with temperature according to an Arrhenius-type behavior. 347SS and 321SS had very little oxidation for 400 and 500 degrees C, indicative of a protective corrosion scale and low corrosion kinetics. Data at 600 degrees C showed that 321SS tended toward linear oxidation behavior based on oxide spallation which was observed on the samples upon removal. Corrosion products at 500 degrees C had phases of iron oxide, with obvious chromium depletion as observed in energy dispersive spectroscopy (EDS) scans. 600 degrees C corrosion layers were primarily iron oxide with obvious phases of sodium ferrite on the outer surface. 680 degrees C marked an excessive rate of corrosion with metal loss in both alloys. Published by Elsevier Ltd.
引用
收藏
页码:878 / 887
页数:10
相关论文
共 17 条
[1]  
Ambrosek J., 2010, 2010 ANS ANN M EMB T
[2]  
[Anonymous], G1032003 ASTM INT
[3]  
[Anonymous], 2002, SAND2002-0120, Appendix P
[4]  
Bradshaw R.W, 1982, SAND828210
[5]  
Bradshaw R.W, 1987, SAND869009
[6]  
Bradshaw R.W, 1980, SAND808856
[7]  
Burgaleta J. I., 2011, SOLARPACES 2011 SOLARPACES 2011
[8]  
Gary J., SOLARPACES2011
[9]  
Gill D.D., 2011, 5 INT C EN SUST, P6
[10]  
Glatzmaier G., WORKSH NEW CONC MAT